OneBalance is led by Stephane Gosselin, a seasoned protocol architect with deep expertise in decentralized infrastructure. From co-founding Flashbots to designing systems at Numerai and Kordina, his expertise spans blockchain research, consensus design, and early-stage protocol development. He’s joined by Daniel Worsley, an experienced operator who previously led operations at Flashbots and scaled platforms like LocalCoinSwap from the ground up. Ankit Chiplunkar, Co-founder and CTO, brings rigorous analytical thinking and technical depth from his PhD in Aerospace Engineering, along with hands-on development experience at Coinbase. Together, Stephane, Daniel and Ankit bring decades of combined experience in blockchain architecture, operational scaling and advanced research and development, making them uniquely positioned to solve Web3’s most persistent UX and interoperability challenges.

At the heart of OneBalance is the concept of the Credible Commitment Machine (CCM), an off-chain trust-minimized system capable of issuing and validating resource locks—cryptographically verifiable user intents that can be executed across chains without fear of equivocation. This novel design solves the core dilemma facing current account models: while EOAs are cheap and fast but not trustworthy (due to equivocation), and Smart Contract Accounts (SCAs) are trustworthy but costly and slow (due to global consensus constraints), Credible Accounts deliver the best of both worlds. The result is an architecture that provides low-latency, gas-abstracted, and composable interactions across any chain—including Ethereum, Solana, and Bitcoin—without requiring custody, bridging, or EOA overhead.

OneBalance enables a “unified account layer” that abstracts the complexity of multichain UX into a single predictable interface. It allows applications and users to aggregate liquidity, enforce programmable permissions, use modern authentication (passkeys, session keys, FIDO), and even outsource execution to solver networks—while retaining the freedom to exit and the ability to choose their security model (TEE, MPC, or chain-native). The implications of this are profound: developers can build intent-based applications like decentralized trading platforms, cross-chain lending protocols, or even non-custodial exchanges that operate with the speed of centralized services, all without losing trust assumptions.

Beyond its architectural breakthrough, OneBalance distinguishes itself through an exceptional developer experience (DX) designed for universal adoption. The OneBalance SDK abstracts away multichain complexity and provides a single, unified API for interacting with tokens, performing swaps, and executing smart contract calls across chains. Developers no longer need to manage different RPC providers, gas payment strategies, or bridge integrations. Instead, they work with predictable, declarative primitives like GetAggregatedBalance, GetQuote, and ExecuteQuote—all chain-abstracted and permissioned through modern authentication methods such as session keys and passkeys.

OneBalance offers support for multiple account configurations (e.g., EIP-4337, EIP-7702, Role-Based), making it compatible with a wide range of existing applications and wallet infrastructure. Developers can choose their security model, whether using smart contract accounts, TEEs, or MPC setups, while retaining full modularity and composability with the broader CAKE working group standards. This flexibility is critical for long-term developer trust and adoption.

The developer tooling is modern, modular, and well-documented—offering built-in support for fast vs. standard execution paths, transaction lifecycle tracking, and real-time fulfillment verification. Importantly, developers building with OneBalance benefit from backwards compatibility with any chain, including non-EVM chains, enabling rich composability and liquidity aggregation in ways previously only possible through custodial solutions.

By making multichain intent execution a drop-in integration rather than a months-long engineering effort, OneBalance becomes not just a protocol, but a platform for the next generation of Web3 apps—whether they be wallets, Telegram bots, NFT marketplaces, or cross-chain DeFi protocols.

For L2IV, OneBalance sits at the convergence of several key theses. It aligns directly with our vision of Web3 infrastructure as programmable, privacy-preserving, and modular, rooted in systems that scale both technologically and socially. The architectural underpinnings of OneBalance—particularly its use of local locks over global consensus, the clean abstraction over user intent, and the emerging standardization from the CAKE Working Group—demonstrate not only technical maturity, but a forward-thinking ethos compatible with our investments in intent-centric protocols, ZK systems, and decentralized coordination primitives.

Moreover, OneBalance’s modular framework and compatibility with EIP-4337 and EIP-7702 reinforces our belief in composable middleware as the path to protocol longevity. The team’s collaborative DNA, with deep engagement from ecosystem builders including Anoma, Ethereum Foundation members, and ZeroDev, reflects our preferred investment profile: not isolated founders, but protocol architects with the capability to unify fragmented efforts under coherent, standards-driven execution.

The fit between OneBalance and L2IV is natural. L2IV provides more than capital—we bring technical stewardship, research amplification, and network acceleration. We see OneBalance not as an application layer product, but as an emerging coordination layer for the multichain economy, capable of abstracting away the seams of blockchain UX in the same way that TCP/IP abstracted the early internet. As this new account abstraction layer takes shape, L2IV is committed to working side by side with the OneBalance team to realize a shared mission: bringing Web3 to its first billion users through scalable, sovereign, and seamless digital infrastructure.

In investing in OneBalance, we are not merely backing a startup—we are backing a shift in worldview. And that is exactly the kind of frontier where L2IV thrives.

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only, and do not constitute an investment recommendation or offer to provide investment advisory services.

Our portfolio company, Luban is building an out of protocol solutions for variety of proposer commitment. The first two iterations are L1 and L2 pre confirmation.

Based sequencing is the critical process of getting transaction data from these rollup chains included periodically into Ethereum's main layer 1 blockchain for security and data availability purposes.

Currently, rollups rely on a centralized sequencer, typically run by the rollup team themselves, to handle this base layer sequencing task. However, this centralized model raises trust issues and could become a bottleneck as rollup usage scales up. So there is a need for a decentralized base layer sequencing solution.

Luban aims to provide this decentralized base sequencing layer through their proposal commitment system. At its core, this system allows various participants, called proposers, to make commitments or proposals about the specific rollup transaction data they intend to include and sequence into future Ethereum L1 blocks.

Other participants, called delegators, can then verify these proposed commitments from proposers and execute the committed actions by actually constructing the layer 1 transactions to sequence the rollup data as proposed. This separates the proposal and execution roles in a decentralized manner.

To make this work, Luban's system has several key components they are building:

1. The proposal commitment scheme itself where proposers stake and declare their intentions to sequence specific rollup data.

2. Pre-confirmation mechanisms to provide probabilistic guarantees in advance about whether committed proposals will likely be executed as scheduled by delegators.

3. Penalty enforcement rules and mechanisms to punish proposers financially if they go back on their committed proposals after staking.

4. Robust communication protocols to efficiently transfer proposal data between all the parties involved - proposers, delegators, verifiers etc.

5. Execution infrastructure and technical capabilities for delegators to reliably execute accepted and verified proposals to sequence rollup data into layer 1.

6. A verification proxy process to cryptographically verify that delegators correctly executed committed proposals as intended, enforcing accountability.

By building this decentralized pre-commitment and execution layer, Luban essentially provides the missing trustless base sequencing piece for rollups to achieve greater security, scalability, and censorship resistance compared to centralized sequencers.

Proposer Commitments

Luban recently posted an article, "Towards a Credible Out-of-Protocol Proposer Commitment Scheme for Ethereum", speculating the industry vertical in detail delving into the concept of proposer commitments within the context of the Ethereum. It explores how these commitments can be utilized to enhance the agency of proposers and generate additional revenue streams without introducing significant operational complexity.

What is it?

A proposer commitment is a mechanism within the Ethereum ecosystem that allows block proposers to make binding promises about their future actions.

At its core, a proposer commitment involves:

1. A willing proposer (or their delegatee) who commits to perform a specific set of actions.

2. A cryptographically signed commitment of these actions.

3. A set of predefined penalties for failing to fulfill the commitment.

4. A verification mechanism to ensure the commitment is met.

Technically, a proposer commitment can be represented as:

Where:

• Sig is the proposer's digital signature

• Hash is a cryptographic hash function

• τ_i represents a series of actions

• φ_i represents the corresponding time or state dependencies for each action

The penalty for failing to meet the commitment can be expressed as:

Where:

• P is the penalty function

• α_t is the historical EVM state at block t

The proposal commitment scheme enables a decentralized two-step process for including rollup transaction data into Ethereum L1 blocks for security and data availability.

Step 1: Commitments from Proposers

• There are "Proposer" participants who stake/commit to including specific rollup transaction data in a future layer 1 block they are assigned to propose

• For example, a Proposer may commit to sequencing rollup block X's data into Ethereum block Y at time Z

• This commitment is made by putting down a stake/bond and broadcasting the commitment via a layer 1 transaction

• Proposers are incentivized to make accurate commitments to earn rewards from their stake

Step 2: Execution by Delegators

• The commitments from Proposers are monitored by "Delegator" participants

• If a Proposer's commitment is verified as valid, Delegators compete to actually construct the layer 1 block at time Z

• The winning Delegator includes the rollup data X in their block Y as per the Proposer's commitment

• Delegators earn a portion of rewards for properly executing committed proposals

This separates the "proposal" and "execution" roles in a decentralized way.

In addition to its proposal commitment system, Luban incorporates another crucial element to enhance the effectiveness and reliability of its decentralized base sequencing solution: the pre-confirmation mechanism. 

This feature works in tandem with the commitment scheme to provide an additional layer of assurance and predictability to the base sequencing process. By offering probabilistic guarantees about the execution of committed proposals before their inclusion in Ethereum's Layer 1, the pre-confirmation mechanism addresses key challenges in transaction finality and user experience. 

Let's explore how this system operates and its significance in Luban's overall architecture.

What’s Been Happening at Around for Pre-confs and Based Sequencing?

There have been different stances in and around pre-confs. Some may be generally accepting of the concept, but their approval may come with specific conditions and concerns.

Firstly, Some might be concerned about the potential impact on Layer 1 consensus. For any L1, the consensus mechanism is crucial for maintaining the security and integrity of the network. With pre-confs, we need to make sure that any implementation of pre-confirmation does not put additional strain on this fundamental layer of the Ethereum network. This could be because overloading the Layer 1 consensus might lead to slower transaction processing times, increased energy consumption, or potential security vulnerabilities.

Secondly, We also need to be cautious about the amount of work required from proposers. In Ethereum, proposers are nodes that suggest new blocks to be added to the blockchain. They play a critical role in the network's operation and security. We do not want proposers to do "too much work" which likely stems from a desire to maintain a balanced and efficient system. 

If proposers are overburdened with additional tasks related to pre-confirmation, it could lead to several potential issues:

1. Increased centralization: If the workload becomes too high, only powerful nodes might be able to serve as proposers, potentially leading to centralization.

2. Reduced participation: A high workload might discourage some nodes from becoming proposers, reducing the overall robustness of the network.

3. Slower block times: If proposers need to perform complex operations for pre-confirmation, it might slow down the block proposal process.

4. Increased vulnerability to attacks: Overworked proposers might be more susceptible to certain types of attacks or manipulation.

We believe there’s potential value in pre-confirmation technology, but it’s not without approaching it with a cautious and pragmatic perspective. 

We are pragmatically more inclined towards the “pricing aspect” of pre-confirmations.

By offering probabilistic guarantees about future block contents and transaction inclusion, pre-confirmation introduces new dynamics that can influence transaction fees, gas pricing, MEV extraction, block space valuation, sequencer compensation, and even lead to the creation of new financial instruments.

Transaction Fees and Gas Pricing

Traditionally, transaction fees are determined in real-time, with users competing for inclusion in the next block. This can result in sudden fee spikes during periods of high network congestion. 

Pre-confirmation, however, allows for a more forward-looking approach to fee determination. It enables a futures market for transaction inclusion, where users could lock in a fee rate for a future block, similar to booking a flight in advance. This approach could lead to more stable and predictable fee structures, as the network can better anticipate and distribute transaction load across future blocks. To implement this, sophisticated smart contracts would be needed to handle these future commitments and adjust fees based on projected network conditions.

Gas pricing, closely tied to transaction fees, could also see a transformation with pre-confirmation. Currently, gas prices fluctuate based on immediate network demand. With pre-confirmation, a more nuanced gas pricing model could emerge. For example, gas prices could be set on a forward curve, much like commodity futures markets. This would allow for more accurate pricing based on expected future network conditions. Implementation would require integration with oracle systems to provide reliable future state predictions and complex algorithms to calculate fair gas prices based on these predictions and historical data.

Impact on Maximal Extractable Value (MEV)

MEV, or Maximal Extractable Value, refers to the value that can be extracted from transaction ordering within blocks. Pre-confirmation systems could make some traditional forms of MEV extraction more challenging by locking in transaction order in advance. However, they could also create new MEV opportunities based on the pre-confirmation process itself. For instance, accurately predicting which transactions will be pre-confirmed could have significant value, allowing traders to position themselves accordingly. This would necessitate the development of new MEV extraction algorithms that consider the pre-confirmation state and potential changes to the consensus mechanism to ensure fair distribution of MEV.

Block Space Valuation

Pre-confirmation introduces a new dimension of time to block space valuation. Block space could be priced differently based on how far in advance it is reserved, similar to how airline seats are priced based on booking time. Implementing this would require developing sophisticated pricing models that consider historical data, future predictions, and the time value of block space. Smart contracts would be essential to handle these time-based reservations and pricing adjustments.

Layer 2 Solutions and Sequencer Compensation

For Layer 2 solutions, pre-confirmation could significantly impact sequencer compensation models. Sequencers could offer pre-confirmation services, guaranteeing transaction inclusion in specific future L2 blocks. This could lead to a more competitive market for sequencing services, with sequencers differentiating themselves based on their ability to provide reliable pre-confirmations. Implementing this would require modifications to existing L2 protocols to incorporate pre-confirmation mechanisms and new economic models to determine fair compensation for sequencers based on their pre-confirmation performance.

New Financial Instruments and Markets

Pre-confirmation opens up possibilities for new financial instruments and markets. We could see the emergence of block space futures or options contracts, where users could purchase the right (but not the obligation) to include a transaction in a specific future block. This could lead to sophisticated hedging strategies for large-scale blockchain users. Market makers specializing in trading these block space derivatives could emerge. 

Is there a Spin to Pre-confs?

Many major L2s have shown interest in pre-confirmations, mainly in areas such as transaction finality, user experience, or interoperability between Layer 2 and Layer 1.

However, regulatory concerns are a significant driver of this interest. In web3, regulatory scrutiny has been increasing, particularly around centralized systems that handle user funds or data. Layer 2 solutions, while built on decentralized networks, often incorporate some centralized components for efficiency.

One of the key centralized components in many Layer 2 systems is the sequencer. A sequencer is responsible for ordering transactions within the Layer 2 network. It collects transactions from users, arranges them in a specific order, and submits them to the Layer 1 network in batches. This centralized ordering allows for faster and cheaper transactions but introduces a single point of control.

The centralized nature of sequencers could potentially attract regulatory attention for several reasons:

1. Control over transaction ordering could be seen as market manipulation, especially in DeFi applications.

2. A central entity having control over user transactions might be viewed as similar to a traditional financial intermediary, potentially subjecting it to banking or money transmission regulations.

3. There might be concerns about data privacy and user information handling.

4. The ability to censor or prioritize certain transactions could be seen as problematic from a regulatory standpoint.

Pre-confirmation technology could potentially offer a way to decentralize the sequencing process. This could work by:

1. Distributing the sequencing role among multiple parties.

2. Implementing a consensus mechanism for transaction ordering.

3. Using cryptographic techniques to ensure fairness and transparency in the ordering process.

By decentralizing sequencers, Layer 2 solutions could potentially address several regulatory concerns:

1. Reduced single point of failure or control, making the system more resilient and less likely to be seen as a centralized intermediary.

2. Increased transparency in transaction ordering, potentially alleviating concerns about market manipulation.

3. Improved censorship resistance, addressing worries about transaction filtering or prioritization.

4. A more distributed responsibility for user data and transaction processing, potentially easing data privacy concerns.

This development also underscores the complex interplay between technological innovation, regulatory landscapes, and growth strategies in this rapidly evolving space.

What is Luban Doing Differently?

With several discussions with the founder of Luban, Harry has made a significant observation about the current state of blockchain development, particularly in the areas of pre-confirmation and Layer 2 scaling solutions. Through his insight, we have identified a crucial gap in the industry's approach to building new features.

Many teams in the blockchain space are heavily focused on building infrastructure. This infrastructure likely includes:

• Pre-confirmation systems

• Decentralized sequencers

• Cross-chain bridges

• Consensus mechanisms

• Data availability solutions

These are all critical components for creating a robust, scalable, and efficient blockchain ecosystem. Teams are investing significant resources into solving complex technical challenges related to scalability, security, and decentralization.

These infrastructure components are undoubtedly critical for creating a robust, scalable, and efficient blockchain ecosystem. Teams are investing substantial time, effort, and resources into solving the complex technical challenges associated with improving scalability, enhancing security, and maintaining decentralization.

While teams are making great strides in developing advanced technical solutions, there is a noticeable lack of attention being paid to how end-users will actually interact with these systems. This gap in focus could lead to several potential issues.

• Firstly, the advanced nature of these technologies might result in user interfaces that are overly complex and difficult for average users to understand and navigate. The intricate technical details that developers are grappling with could inadvertently be reflected in the user interface, creating a barrier for non-technical users.

• Secondly, users might need to understand complex concepts like Based sequencing & pre-confirmation to effectively use these systems. This creates a significant technical barrier that could deter many potential users who are not well-versed in blockchain technology.

• Thirdly, in the rush to solve technical problems, considerations of UX/UI design might be overlooked. This could result in systems that, while technically impressive, are not user-friendly or intuitive to interact with.

• Lastly, without a proper focus on user interaction, there might be insufficient resources dedicated to educating users on how to leverage these new technologies. This could create an education gap, where even if the technology is powerful and user-friendly, users may not understand how to take advantage of its capabilities.

We see this gap in focus as a significant opportunity for differentiation. This could catalyze adoption, bridging the gap between advanced infrastructure and mainstream users who may be intimidated by the complexity of blockchain technology.

In a space where many projects have similar technical capabilities, superior user interaction could be a key differentiator.

Conclusion

We believe that Luban with its pre-confirmation mechanism and a strategic focus on delivering an exceptional user experience is an excellent approach to take.

This two-pronged approach tackles both the technical challenges of decentralized base layer sequencing and the oft-neglected aspect of user accessibility. In an ecosystem where many projects offer similar technical capabilities, Luban's emphasis on seamless user interactions could prove to be a powerful differentiator, unlocking the potential for their cutting-edge solutions to reshape how users interact with L2s and the future of Ethereum's roadmap towards a scalable, secure, and decentralized web3 ecosystem.

Find L2IV at l2iterative.com and on Twitter @l2iterative

Subscribe now

Author: Arhat Bhagwatkar, Research Analyst, L2IV (@0xArhat)

References

• Towards a Credible Out-of-Protocol Proposer Commitment Scheme for Ethereum - Anatomy of Proposer Commitments

• Ethereum Foundation: Sequencing and Preconfirmations Call

• Based Preconfirmations

• Preconfirmations: The Fulfillment-Delivery Paradigm

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only and do not constitute an investment recommendation or offer to provide investment advisory services.

Aori is a decentralized infrastructure layer for high-frequency intent execution across the rapidly evolving DeFi ecosystem.

At its core, Aori has developed proprietary "flash orderbook" technology that can universally aggregate liquidity from automated market makers, orderflow auctions, solver systems and more into a unified, ultra-low latency liquidity pool. This unified liquidity source spans across all major blockchain networks and rollup solutions.

However, Aori goes far beyond just facilitating high-frequency trading and market making strategies. Their key innovation is providing institutional-grade infrastructure to enable the seamless and real-time execution of any operational or compositional "intent" that requires securely tapping into DeFi's native liquidity.

Some of the diverse "intents" that Aori's architecture empowers include arbitrage across fragmented DeFi markets, liquidating undercollateralized positions, rebalancing liquidity pools or fund portfolios, sourcing liquidity for large trades or concentrated positions, extracting maximal value from MEV opportunities, and facilitating high-speed request-for-quote (RFQ) flows.

Note: We did a deep dive on Order Flow Auctions (OFAs) where you can dive deep into the why and how of OFAs in detail

Aori achieves this unparalleled versatility through an innovative model that incentivizes and integrates various specialized participants like elite MEV searchers, solvers, market makers and liquidity providers. These diverse providers supply liquidity into Aori's unified orderbook, allowing any "intent" to be instantly matched and executed at ultra-competitive rates.

Rather than being a single monolithic platform, Aori's architecture embraces composability at its core. It seamlessly integrates best-of-breed execution venues, novel DeFi primitives, and innovative liquidity sources as they emerge. This positions Aori as the essential integration fabric stitching together the entire rapidly evolving DeFi trading and operational stack.

Aori’s Architecture

Aori's off-chain, intent-centric matching layer is a fundamental component of its infrastructure, designed to facilitate high-frequency trading and efficient order matching in the DeFi ecosystem. This layer operates independently from the underlying blockchain, allowing it to process and match trading intents without the need for immediate on-chain execution. By moving the matching process off-chain, Aori overcomes the limitations of purely on-chain solutions, such as high latency and gas costs, which are particularly detrimental to high-frequency trading strategies.

The matching layer functions as a central hub for receiving and processing trading intents from various market participants, including market makers, DeFi protocols, and individual traders. These intents are expressed using a standardized format that captures the desired asset pair, quantity, price, and any specific conditions or constraints. The standardization of intent formats is crucial for ensuring seamless compatibility and efficient processing by the matching layer.

Upon receiving the trading intents, Aori's market makers and solvers within the matching layer analyze them to find the most optimal matches. This process takes into account a wide range of factors, such as price, quantity, available liquidity, and any specified conditions. The matching layer leverages Aori's liquidity aggregation capabilities to tap into liquidity pools across multiple DEXs and AMMs, ensuring that users receive the best possible execution prices for their trades.

One of the key advantages of Aori's matching layer is its ability to accommodate complex trading strategies and conditions. Users can express intents for multi-leg trades, limit orders, and conditional orders, providing them with the flexibility to adapt to changing market dynamics and execute advanced trading strategies. This level of flexibility is unparalleled in the DeFi space and sets Aori apart from traditional AMMs and order book-based systems.

Once the matching layer finds optimal matches for the trading intents, it generates a trade execution plan. This plan outlines the specific steps required to complete the trade, but allows solvers and market makers to compose arbitrary interactions with any external DeFi protocol or liquidity source. The market maker or solver then puts this through Aori’s modular settlement layer, individual contracts on deployed chains, which interacts with the relevant on-chain protocols and smart contracts to finalize the trade. By separating the matching process from the on-chain settlement, Aori can achieve higher throughput and lower latency compared to purely on-chain solutions.

The off-chain nature of Aori's matching layer offers several significant benefits for high-frequency trading. 

• Firstly, it drastically reduces the latency associated with on-chain transactions. High-frequency trading strategies often rely on rapid execution and minimal delays to take advantage of fleeting market opportunities. By processing and matching trading intents off-chain, Aori can provide near-instant trade confirmation and execution, enabling traders to capitalize on these opportunities effectively.

• Secondly, off-chain matching helps to minimize the transaction costs associated with high-frequency trading. In a purely on-chain environment, each individual trade would incur gas costs, which can quickly accumulate and erode the profitability of high-frequency strategies. Aori's matching layer mitigates this issue by batching multiple trades together and settling them on-chain in a single transaction. This batching mechanism significantly reduces the overall gas costs, making high-frequency trading more cost-effective and accessible to a wider range of market participants.

• Lastly, market makers and solvers achieve significantly more confidence trading through the matching confirmations that Aori gives to allow them to hedge or quantify their risk. Through this, more sophisticated market making strategies can be used to offer users more competitive prices.

Moreover, Aori's off-chain matching layer can process a substantially higher volume of trading intents compared to purely on-chain solutions. This increased throughput is essential for accommodating the demands of high-frequency trading, where a large number of trades need to be executed within short timeframes. By leveraging off-chain processing, Aori can handle a higher influx of trading intents without causing congestion or slowdowns on the underlying blockchain network.

Another significant benefit of Aori's off-chain matching layer is its ability to aggregate liquidity from multiple sources. In the DeFi ecosystem, liquidity is often fragmented across various DEXs and AMMs, which can lead to suboptimal execution prices and increased slippage. Aori's matching layer addresses this issue by tapping into liquidity pools across multiple protocols, effectively consolidating the available liquidity. This liquidity aggregation ensures that high-frequency traders can access the best possible prices and minimize slippage, even for large-volume trades.

The combination of reduced latency, lower transaction costs, increased throughput, and improved liquidity aggregation makes Aori's off-chain matching layer a powerful tool for high-frequency trading in the DeFi space. By providing a fast, efficient, and cost-effective means of executing trades, Aori enables market makers, trading firms, and other sophisticated players to deploy their high-frequency strategies effectively. This, in turn, contributes to increased liquidity, tighter spreads, and improved overall market efficiency within the DeFi ecosystem.

Furthermore, Aori's off-chain matching layer serves as a foundation for the development of advanced trading tools and platforms. By offering a reliable and high-performance infrastructure for matching trading intents, Aori enables the creation of sophisticated trading algorithms, quantitative strategies, and automated trading systems. These tools can leverage Aori's intent-based framework and liquidity aggregation capabilities to execute complex strategies and adapt to rapidly changing market conditions.

The off-chain matching layer also plays a crucial role in enabling cross-protocol interactions and facilitating the composability of DeFi protocols. By abstracting away the complexities of interacting with multiple on-chain protocols and providing a standardized interface for expressing trading intents, Aori makes it easier for developers to build applications that span across different DeFi ecosystems. This unlocks new possibilities for creating innovative financial products and services that combine the functionalities of various DeFi building blocks.

We have visualized the architecture of Aori's off-chain, intent-centric matching layer:

1. Market participants submitting trading intents to Aori's off-chain matching layer.

2. The matching layer standardizing the intent formats for efficient processing.

3. Market makers and solvers analyzing the intents, considering factors such as price, quantity, and conditions.

4. Liquidity aggregation from multiple DEXs and AMMs.

5. Finding optimal matches based on the analyzed intents and aggregated liquidity.

6. Calldata is generated for the matched market maker or solver to settle this on-chain and fulfill the intent.

7. Passing the trade execution plan to one of Aori’s modular settlement contracts.

8. The settlement layer initiating token transfers and swaps, interacting with DeFi protocols, monitoring on-chain transactions, and updating internal records.

9. Finalizing and confirming the executed trades.

10. Updating user balances and records based on the completed trades.

Aori's Intent-Based Framework

Intent-based trading is an intricate paradigm that focuses on capturing and executing the underlying intentions behind a trade rather than merely the specific parameters of an order. In traditional order book-based systems, users typically specify the exact details of their desired trade, such as the asset pair, quantity, price, and order type (e.g., limit, market). While this approach works well for simple spot trades, it can be limiting when dealing with more complex trading strategies or when interacting with various DeFi protocols.

Aori's intent-based framework, on the other hand, allows users to express their trading intentions in a more flexible and adaptable manner. Instead of rigidly defining the trade parameters, users can convey their desired outcomes or objectives along with any specific conditions or constraints. For instance, a user could express an intent to swap a certain amount of token A for token B, but only if the price is within a specific range and the slippage is below a certain threshold.

To facilitate this intent-based approach, Aori employs a standardized format for expressing trading intents. This format includes fields such as the intent type (e.g., swap, provide liquidity), asset pairs, quantities, price ranges, slippage tolerances, and any other relevant conditions. By adopting this standardized format, Aori enables seamless communication between users, liquidity providers, and the underlying matching engine.

Aori is UniswapX With No Limits

Aori can be described as "UniswapX with no limits," allowing any kind of trading based on its intent-based framework. This approach offers several key advantages over traditional AMMs like Uniswap:

1. Flexible trading strategies: For example, a user could express an intent to swap token A for token B, but only if the price of token B is within a specific range and the slippage is below a certain threshold. This flexibility enables traders to execute sophisticated strategies that are not possible on traditional AMMs like Uniswap.

2. Optimal execution and reduced slippage: By tapping into liquidity across multiple DEXs and AMMs, Aori can find the most efficient execution path for a given trade, splitting it across different pools and protocols if necessary through Aori's off-chain matching layer and liquidity aggregation capabilities. This optimized execution reduces the price impact of large trades and helps traders achieve better overall returns.

3. Capital efficiency for liquidity providers: Liquidity providers can express their intent to provide liquidity with specific parameters, such as the desired asset pair, price range, and fee tier. Aori's matching layer then optimally matches these liquidity provision intents with trading intents, ensuring that liquidity is allocated where it is needed most. This dynamic liquidity allocation improves capital efficiency for liquidity providers and reduces the opportunity cost of holding idle assets.

4. Cross-protocol interactions: For example, a user could express an intent to borrow an asset from a lending protocol, swap it for another asset on Aori, and then use the resulting asset as collateral in a yield farming strategy on another protocol. Aori's matching layer handles the necessary cross-protocol interactions and ensures that the entire trade is executed atomically, minimizing the risks associated with manual, multi-step transactions.

5. Extensibility and customization: By defining new intent types and integrating with external protocols, developers can build innovative applications that go beyond the limitations of traditional AMMs. This extensibility fosters a vibrant ecosystem of DeFi products and services that can cater to the diverse needs of traders and investors.

Handling Complex Intents: Perps and Spot Tokens

While Aori's intent-based framework is designed to support a wide range of trading strategies, some intents can be particularly complex and require special handling. 

Two examples of such intents are those involving 

• perpetual contracts (perps): Handling intents related to perps can be challenging due to funding rate payments and the need for continuous price updates. 

• Low Capital Spot tokens: While spot token trades may seem straightforward, handling intents related to spot tokens can also be complex, particularly when dealing with low-liquidity tokens or tokens with unique trading pairs.

To effectively handle these complex intents, Aori's infrastructure can effectively:

1. Aggregate liquidity from multiple sources to ensure sufficient depth and minimize slippage.

2. Provide real-time price updates and funding payment calculations for perps.

3. Support a wide range of token pairs and trading pairs, including less common or newly listed tokens.

4. Implement robust risk management mechanisms to handle the volatility and potential liquidations associated with perps.

5. Optimize gas costs and transaction fees for spot token trades, especially during periods of high network congestion.

By addressing these challenges and providing a seamless trading experience for complex intents, Aori demonstrates its versatility and ability to support the diverse needs of the DeFi trading ecosystem.

Build with Aori: Composable Products and Services

Now that we know that Aori's composable infrastructure is a key aspect of its value proposition, this enables a wide range of products and services to be built on top of its foundation. 

Composability, in the context of DeFi, refers to the ability of different protocols and services to seamlessly interact and build upon each other, creating a vibrant ecosystem of interoperable financial applications. Aori's architecture is designed to support and enhance this composability, allowing developers to create innovative and complex financial products that leverage the capabilities of multiple DeFi building blocks. 

Let’s explore a few.

Real-World Applications and Use Cases 

CEX-DEX arbitrage

One of the most compelling use cases for Aori's infrastructure is CEX-DEX arbitrage. This end-to-end flow demonstrates the power of Aori's order book and its ability to facilitate high-frequency trading strategies that capitalize on price discrepancies between CEXs and DEXs.

In this scenario, market makers continuously monitor the prices of assets on both Aori's order book and various CEXs. Aori's order book aggregates pricing information from multiple DEXs through its network of solvers, providing a comprehensive view of the DeFi market.

When a market maker identifies a price discrepancy between Aori's order book and a CEX, they can execute an arbitrage trade. For example, if the price of ETH is lower on Aori's order book compared to a CEX, the market maker can buy ETH on Aori and simultaneously sell it on the CEX, profiting from the price difference.

To complete the arbitrage trade, the market maker fills orders on both sides - buying on Aori and selling on the CEX. Aori's off-chain matching engine and solver network ensure that the orders are filled at the best available prices across multiple DEXs. Once the trade is executed, the market maker can hedge their position to lock in the profit and manage risk.

This CEX-DEX arbitrage flow showcases Aori's ability to provide a seamless and efficient trading experience, combining the liquidity of multiple DEXs with the speed and flexibility of off-chain order matching. By facilitating such arbitrage opportunities, Aori helps to improve market efficiency, reduce price discrepancies, and create a more cohesive trading ecosystem.

And a few examples

Aori's composable architecture unlocks the potential for a diverse ecosystem of innovative financial products and services to be built on its infrastructure. By leveraging Aori's intent-based framework, off-chain matching layer, and liquidity aggregation capabilities, developers can create solutions that seamlessly combine functionalities from multiple DeFi building blocks.

Routing-as-a-Service 

Efficient trade routing is critical for optimal execution and minimizing slippage, especially for complex multi-asset or cross-protocol trades. However, building robust routing infrastructure requires significant resources and expertise, posing challenges for individual DeFi protocols.

This routing-as-a-service offering addresses this by allowing protocols to outsource their routing needs. Protocols express routing intents through Aori's standardized format, specifying desired assets, quantities, and constraints. Aori's matching layer then finds the optimal cross-protocol execution path, leveraging its aggregated liquidity network. The settlement layer executes the trade plan efficiently, abstracting away routing complexities.

Structured Products

Aori enables the creation of structured financial instruments tailored to specific risk profiles and investment objectives. Protocols can design complex products like yield-enhanced tokens, risk-hedged portfolios, or multi-asset baskets by expressing intricate trading intents.

For instance, a structured product could combine yield-generating tokens from various lending protocols, hedged with derivatives for risk mitigation. Aori's matching layer handles the underlying component interactions, swaps, and protocol integrations required to construct and manage these instruments dynamically.

Yield Optimization Strategies 

Navigating DeFi's fragmented yield opportunities can be daunting for individual users. Aori simplifies this by allowing protocols to build automated yield optimization strategies as complex trading intents.

These intents capture desired assets, yield targets, risk tolerances, and other parameters. Aori's matching layer analyzes the intents and generates an optimal cross-protocol yield farming plan, continuously adjusted based on changing market conditions. This enables sophisticated, user-friendly yield optimization products without manual effort.

Arbitrage Bots

Aori's high-frequency trading capabilities make it ideal for building arbitrage bots that exploit fleeting price discrepancies across DEXs and AMMs. Protocols express arbitrage intents, specifying assets, constraints, and conditions like accessing flash loans.

Aori's matching layer executes these intents efficiently, splitting trades across liquidity sources for optimal pricing. This seamless integration with lending protocols, derivatives platforms, and liquidity fragmentation enables creating adaptive arbitrage strategies that capture short-lived opportunities.

Portfolio Rebalancing and Asset Management 

Aori can power advanced portfolio rebalancing and asset management solutions. Protocols build intricate rebalancing intents based on risk parameters, portfolio targets, and yield strategies. The intent framework coordinates actions across lending, yield farming, arbitrage, and hedging opportunities.

For example, a product could automate rebalancing while compounding yields from liquidity providing and capturing arbitrage. Aori's architectural composability enables bundling these capabilities into unified asset management experiences.

From capital-efficient routing and tailored structured products to automated yield harvesting, arbitrage bots, and holistic portfolio solutions, Aori's architecture empowers developers to compose DeFi's building blocks into an ecosystem of powerful and user-friendly financial products and services.

Aori's Approach to MEV on Rollups

As the DeFi ecosystem continues to evolve and embrace rollup solutions for scalability, the challenges and considerations surrounding Maximal Extractable Value (MEV) on Layer 2 have come to the forefront.

Rollups face unique security and economic risks that differ from those encountered on Layer 1. These risks arise from the unique architecture and design choices of rollups, which differ from those of the base layer. Understanding these risks is crucial for developing effective solutions and ensuring the long-term sustainability of rollup-based systems.

Security Risks in Rollups

While rollups inherit some of the security properties of Layer 1, such as reorg resistance, data availability, and validity, they also introduce new challenges and risks that must be addressed.

1. Real-Time Censorship: In a rollup, the operator (or a small set of operators) is responsible for sequencing transactions and submitting them to the base layer. This concentration of power creates the risk of real-time censorship, where the operator may exclude certain transactions or prioritize their own transactions over those of other users. Real-time censorship can lead to unfair transaction ordering, denial of service, and the violation of users' expectations of inclusivity and fairness.

2. Data Availability Challenges: Rollups rely on the base layer for data availability, meaning that the data required to reconstruct the rollup state must be accessible on Layer 1. If the data becomes unavailable or is withheld by malicious actors, it can compromise the integrity and security of the rollup system. Ensuring reliable data availability is crucial for maintaining the trustworthiness and resilience of rollups.

3. Validator Collusion: In some rollup designs, a small set of validators is responsible for verifying and attesting to the validity of rollup transactions. If these validators collude or are compromised, they can potentially approve invalid transactions or manipulate the rollup state. Mitigating the risk of validator collusion requires careful design choices, such as staking mechanisms, slashing penalties, and robust fraud proofs.

Economic Equity Risks in Rollups

"Economic equity" in rollups, which refers to the fair distribution of economic benefits and costs among all participants in the system. Economic equity risks arise when certain parties, such as operators or large stakeholders, have the ability to capture a disproportionate share of the value generated by the rollup, at the expense of other users.

1. Rent Extraction: Rollup operators may have the ability to extract excessive fees or capture a disproportionate share of the value generated by the system. This rent extraction can occur through the manipulation of transaction fees, the prioritization of operator-controlled transactions, or the exploitation of MEV opportunities. Rent extraction undermines the economic fairness and sustainability of the rollup system, as it diverts value away from users and towards a centralized entity.

2. MEV Exploitation: MEV can be exploited by operators, searchers, or other privileged actors who have access to pending transaction information. MEV exploitation can lead to front-running, back-running, and other forms of value extraction that disadvantage ordinary users and undermine the integrity of the system.

3. Centralization of Wealth: Rollups may exhibit a tendency towards the centralization of wealth, where a small number of large stakeholders or operators capture a disproportionate share of the economic benefits. This centralization can occur through the accumulation of transaction fees, MEV profits, or the control over key infrastructure components. Centralization of wealth undermines the principles of economic equity and fairness, as it concentrates power and influence in the hands of a few actors.

Aori's Approach to Addressing Security and Economic Equity Risks

Aori tackles these distinct rollup challenges head-on with its innovative infrastructure layer for high-frequency intent execution. At its core is a proprietary intent-based framework combined with an off-chain matching layer.

Users express trading intents with specific conditions rather than explicit on-chain orders. These intents are then matched off-chain against aggregated liquidity from DEXs, AMMs, solvers, and market makers before seamlessly executing on the respective rollup.

This architecture mitigates several key risks:

• Real-Time Censorship Resistance: By obfuscating transactions until the final execution stage, Aori's intent model prevents rollup operators from arbitrarily censoring or re-ordering users' activities in real-time based on their economic incentives. The off-chain matching also distributes this responsibility across specialized solvers and liquidity providers.

• Rent Extraction Prevention: Aori's composability allows integrating diverse liquidity sources, minimizing any single party's ability to extract excessive rents through fee manipulation or liquidity monopolization. Its governance mechanisms and service agreements foster accountability for fair operation.

• Economic Equity and MEV Mitigation: The intent framework levels the playing field by concealing transactions from public mempool monitoring, preventing front-running and MEV exploitation by privileged miners/validators. The separation of roles and participation incentives promote an equitable distribution of MEV profits and prevent excessive centralization.

Aori's architecture is purposefully extensible, allowing it to adapt its feature set as new MEV strategies and economic vulnerabilities arise in the rollup environment over time.

Aori’s Competitive Moat

In the rapidly evolving DeFi landscape, Aori's unique approach to high-frequency trading and intent-based execution sets it apart from potential competitors. We are confident that Aori’s competitors would face significant challenges in attempting to verticalize on specific use cases or replicate Aori's infrastructure. But that does not delimit the risk of competition for Aori.

Architectural Superiority

As Aori gains traction and demonstrates the value of its intent-based trading infrastructure, it may attract imitators seeking to replicate its model. Some competitors may attempt to build vertically-focused solutions that target specific use cases, such as liquidations or rebalancing. These narrow solutions could potentially offer optimized performance for their specific use case, posing a risk to Aori's market share in those areas.

However, we believe and was also confirmed by the founders of Aori that the competitors would face significant challenges in replicating Aori's infrastructure and advantages. Building a high-performance, intent-based trading system requires substantial technical expertise, resources, and time. Competitors would need to invest heavily in developing their own off-chain matching engines, liquidity aggregation mechanisms, and integration layers. Additionally, they would need to establish relationships with sequencers and other key stakeholders in the DeFi ecosystem, which can be difficult and time-consuming.

Unparalleled Execution Efficiency 

Also, Aori's execution efficiency and co-location with sequencers provide a significant competitive advantage. Aori's off-chain matching engine and solver network are designed to minimize the impact of network congestion and gas costs, ensuring that trades are executed quickly and efficiently. By co-locating its infrastructure with sequencers, Aori can ensure that its trades are processed and confirmed rapidly, reducing the potential for front-running or other forms of market manipulation. Competitors would struggle to match Aori's execution efficiency without significant investments in infrastructure and sequencer relationships.

Execution speed and reliability are critical factors in DeFi trading, particularly for high-frequency strategies. Traders need to be able to take advantage of fleeting market opportunities and execute complex strategies without being impacted by network congestion or other performance issues. Aori's focus on execution efficiency and co-location with sequencers ensures that its users can trade with confidence, knowing that their orders will be filled quickly and reliably. This is a significant advantage over competitors who may struggle with slower or less reliable execution.

Liquidity Depth and Pricing Power

Aori's ability to aggregate liquidity from multiple sources, including DEXs and AMMs, is another key competitive advantage. By pooling liquidity from across the DeFi ecosystem, Aori can provide its users with deeper order books and better prices, even for less liquid trading pairs. This is particularly important for high-frequency trading strategies, which often rely on the ability to quickly execute large orders without significant slippage. Competitors who focus on a single liquidity source or a narrow range of trading pairs may struggle to match Aori's liquidity and pricing.

As Aori continues to grow and integrate with more DeFi protocols, it benefits from powerful network effects. Each new integration and partnership increases the liquidity and trading volume on the platform, attracting more users and further enhancing its liquidity and pricing. Competitors would need to overcome these network effects by building their own ecosystem of integrations and partnerships, which is a significant challenge. They would need to convince protocols and users to integrate with their platform instead of Aori, which may be difficult given Aori's established presence and growing network effects.

Aori’s Focus on Atomic Financial Intents (Asset Swaps)

Aori's focus on perfecting the user experience for atomic financial intents, particularly asset swaps, is another key competitive advantage. By providing a seamless, intuitive, and efficient interface for executing swaps, Aori can attract and retain users who value simplicity and ease of use. This is particularly important for less technical users or those who are new to DeFi trading, as it lowers the barriers to entry and makes it easier for them to participate in the ecosystem.

Conclusion

At L2IV, we are proud backers of Aori for being the Infrastructure layer for high-frequency intent execution. By providing a high-performance, composable, and MEV-resistant platform, Aori enables users to execute complex trading strategies and access deep liquidity across multiple DEXs and AMMs.

As DeFi continues rapidly evolving, institutional adoption will be pivotal for unlocking its potential to transform global markets. However, to onboard institutional pioneers at scale, a robust infrastructure layer must emerge to navigate DeFi's complexities. Aori stands poised as this institutional infrastructure - providing the high-performance plumbing to facilitate seamless high-frequency execution and choreography of any operational or trading "intent" across DeFi's fragmented landscape.

Through its proprietary flash orderbook, off-chain intent-matching engine, and composable integration of diverse liquidity sources, Aori consolidates DeFi's disparate liquidity pools into one unified low-latency layer. This empowers institutional users to tap into available market depth without sacrificing speed or liquidity. Crucially, Aori transcends the limitations of conventional platforms by enabling real-time execution of any intent requiring access to native DeFi liquidity and protocols. From cross-market arbitrage and liquidations to rebalancing vaults and sourcing liquidity, Aori accelerates these high-frequency workflows.

Obfuscating transactions, integrating diverse participants as liquidity providers, and distributing roles across specialized solvers mitigate emerging risks surrounding MEV on rollup ecosystems. 

Most importantly, Aori catalyzes an interoperable ecosystem where institutional traders, funds, and fintech builders can seamlessly access and choreograph DeFi's operational possibilities at industrial scale through composable offerings. From outsourced routing services and tailored structured products to automated yield harvesting and holistic portfolio engines, Aori empowers institutions to choreograph DeFi's possibilities uninhibited by monolithic limitations.

About Aori

• Website

• Documentation

We would like to express our sincere gratitude to Joshua, Hilliam and the Aori entire team for their invaluable insights into this article

Find L2IV at l2iterative.com and on Twitter @l2iterative

Subscribe now

Author: Arhat Bhagwatkar, Research Analyst, L2IV (@0xArhat)

References

• The Future of On-Chain Liquidity

• Exploration of MEV Latencies

• Latency and the Last Aggregator Problem

• The High-Frequency Trading Arms Race

• Quantifying the High-Frequency Trading “Arms Race”

• Rolling in the Shadows - Analysing the Extraction of MEV Across Layer-2 Rollups

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only and do not constitute an investment recommendation or offer to provide investment advisory services.

Aligned Layer is being built as a universal zero-knowledge (ZK) proof verification layer. Its sole purpose is to provide cost-effective and fast verification of ZK proofs from various proving systems, which will benefit Ethereum by improving its verification capabilities.

Here’s our deep dive.

Introduction

Aligned Layer is a decentralized verification layer designed to provide fast, affordable, and scalable verification of zero-knowledge proofs, serving as a universal source of truth for the Ethereum ecosystem and beyond. It leverages Ethereum's security and decentralization through restaking via EigenLayer while offering a more efficient and flexible alternative to direct on-chain verification.

The primary purpose of Aligned Layer is to address the high costs and inefficiencies associated with verifying ZK proofs on Ethereum. Currently, verifying ZK proofs on Ethereum can be expensive and time-consuming, especially for certain proving systems. This creates an unfair market situation where some proofs are more costly to verify than others, leading to barriers for projects and developers who want to integrate ZK solutions into their applications.

LambdaClass, the incubation team behind Aligned Layer, recognized this problem and set out to create a universal verification layer as a dedicated and optimized infrastructure for proof verification. The goal is to provide a cost-effective, fast, and scalable solution that benefits Ethereum and other protocols. By reducing verification costs, improving efficiency, and enabling faster finality for ZK-based applications, Aligned Layer aims to unlock the full potential of ZK technology and foster a thriving ecosystem of ZK-powered solutions. Let’s explore the problem in more detail.

Problem

The main problem that Aligned Layer is addressing is the high cost and inefficiency of verifying ZK proofs on Ethereum. This issue arises from several factors:

1. Expensive verification on Ethereum: Currently, verifying ZK proofs on Ethereum is costly for many proof systems. This is because the verification process requires significant computational resources and gas fees. As a result, some proof systems are more expensive to verify than others, creating an unfair market situation. 

2. Lack of a universal verification layer: Ethereum currently lacks a dedicated and optimized layer for verifying ZK proofs from different proving systems. Each proof system has its own verification requirements and costs, leading to fragmentation and complexity. 

3. Latency issues caused by batch verification: Teams often resort to combining multiple proofs over a period of time and verifying them in a batch on Ethereum to amortize the high verification costs. However, this approach introduces latency, as it takes time to accumulate enough proofs and then combine them together.

4. Limited scalability and high gas costs: Ethereum's current architecture and gas model pose challenges for the widespread adoption of ZK proofs. As more projects and users rely on ZK technology, the demand for verification increases, leading to higher gas costs and potential scalability issues. 

Technically, Aligned Layer tackles these problems by:

1. Utilizing a separate verification network: Aligned creates a dedicated network specifically designed for verifying ZK proofs. This network consists of operators who run the verification algorithms and reach consensus on the validity of the proofs. By offloading the verification process to a specialized network, Aligned reduces the burden on Ethereum and enables faster and cheaper verification.

2. Supporting various proof systems: Aligned is designed to be compatible with a wide range of ZK proof systems, such as Plonk, STARKs, Groth16, and more. This universality allows projects and developers to choose the most suitable proof system for their use case without worrying about the underlying verification infrastructure. Aligned provides a standardized interface for submitting and verifying proofs, simplifying the integration process.

3. Recursive proof composition: Aligned Layer introduces an aggregation mode that employs recursive proof composition techniques. By using proof systems like Halo2 or Risc0, Aligned Layer can efficiently aggregate multiple verified proofs into a single proof. This aggregated proof can then be verified on Ethereum, reducing the overall verification cost and latency. The recursive composition allows for a more compact and efficient representation of the proofs, making it easier to scale.

4. Integration with EigenLayer and Support for zkEVM: Aligned Layer is designed to work seamlessly with zkEVM. By leveraging the security and infrastructure provided by zkEVM, Aligned Layer benefits from features like restaking and objective slashing. Restaking allows operators to use their staked ETH to participate in the consensus and security of Aligned Layer, while objective slashing enables the slashing of operators who report incorrect verification results. This integration ensures the security and reliability of the verification process.

Aligned Layer offers two distinct modes of verification to cater for different use cases and requirements: fast verification mode and aggregation mode. The choice between fast verification mode and aggregation mode depends on the specific requirements of the ZK-based application. 

• Applications that prioritize real-time responses and low latency can benefit from fast verification mode, while 

• those that aim to minimize verification costs and handle a high volume of proofs can leverage aggregation mode. 

Aligned Layer provides the flexibility to choose the most suitable verification mode based on the needs of the application, enabling developers to optimize their ZK-based solutions effectively.

Aligned Layer Architecture

The architecture of Aligned Layer is designed to facilitate the verification of proofs generated using various proving systems, such as Plonk, STARKs, Groth16, and more. It is built to receive these proofs from different clients, store them securely, and enable efficient verification by a network of operators.

At its core, Aligned Layer is a system that allows clients, which can be individuals, projects, or applications, to submit proofs for verification. These proofs can be generated using different proving systems, each with its own unique properties and characteristics. For example, 

• Plonk proofs are known for their short proof sizes and fast verification times.

• STARKs offer transparency and scalability. 

• Groth16, on the other hand, is widely used for its efficiency and compatibility with various applications.

When a client submits a proof to Aligned Layer, the proof is first stored in a dedicated data availability (DA) layer. Once a proof is stored in the DA layer, it becomes available for verification by the network operators. Aligned Layer relies on a decentralized network of operators who are responsible for verifying the submitted proofs. These operators are typically individuals or entities that have a stake in the network and are incentivized to perform the verification work accurately and efficiently.

Note: EigenDA is the default DA Layer, but Aligned Layer can integrate with Celestia and Avail as well.

The process of assigning proofs to operators for verification is based on a predefined allocation mechanism. This mechanism ensures that each operator is assigned a specific proof to verify, preventing duplication of work and ensuring a fair distribution of verification tasks among the operators. When an operator is assigned a proof to verify, they retrieve the proof from the DA layer. The retrieval process is designed to be efficient and secure, allowing the operator to access the proof without any unnecessary delays or security risks.

Once the operator has retrieved the proof, they run the corresponding verification algorithm associated with the proving system used for that particular proof. Each proving system has its own verification algorithm, which is designed to check the validity and correctness of the proof. For example, the verification algorithm for Plonk proofs would be different from the one used for STARKs or Groth16 proofs.

The verification algorithm performs the necessary computations and checks to determine whether the proof is valid and meets the required criteria. This process involves mathematical calculations and cryptographic operations specific to the proving system. The operator executes these computations to ensure that the proof is sound and has not been tampered with.

After all the operators have completed the verification process for their assigned proofs, they engage in a consensus mechanism to reach an agreement on the validity of each proof. Once a consensus is reached, the verification results are posted to an Ethereum smart contract. The smart contract acts as a tamper-proof and publicly verifiable record of the proof verification outcomes. By posting the results on the Ethereum blockchain, Aligned Layer leverages the security and immutability provided by Ethereum.

The integration with Ethereum through the smart contract serves multiple purposes. 

• First, it provides a trustless and transparent way to track and verify the proof verification process. Anyone can access the smart contract and view the verification results, ensuring transparency and accountability.

• Second, posting the verification results on Ethereum allows other applications and protocols to rely on the verified proofs. They can interact with the smart contract to check the validity of proofs and make decisions based on the verification outcomes. This enables seamless integration and interoperability between Aligned Layer and other systems built on Ethereum.

Proof Aggregation

While posting the verification results on Ethereum provides transparency and interoperability, verifying every single proof on Ethereum can be inefficient and costly, especially as the number of proofs grows. To address this challenge, Aligned Layer introduces an aggregator component that performs proof aggregation.

The diagram above illustrates the proof aggregation process within the Aligned Layer architecture, showcasing how different types of ZK proofs are efficiently aggregated and posted on the Ethereum blockchain.

Proofs generated by different proving systems, such as STARKs, Plonk, and Groth16, are passed to the general prover component, which consists of specialized verifiers for each proof type. These verifiers validate the proofs and then aggregate them using recursive verification techniques.

The aggregation process involves using general-purpose provers like Halo2 or Risc0 to convert proofs into a consistent format (e.g., a Risc0 STARK proof) and then performing recursive verification. Recursive proof composition is a powerful concept that allows for the efficient combination of multiple proofs into a single, compact proof.

The aggregator utilizes advanced cryptographic techniques and proof systems, such as Halo2, Risc0, Groth16, or Plonk, to achieve recursive proof composition. These proof systems allow for the efficient combination of multiple proofs into a single, compact proof while preserving their validity and security properties. By aggregating the proofs, Aligned Layer significantly reduces the computational and storage overhead on the Ethereum network, making the verification process more scalable and cost-effective.

Instead of verifying each proof separately on Ethereum, only the aggregated proof needs to be verified. This aggregated proof represents the validity of all the individual proofs that were verified by Aligned Layer. By verifying the aggregated proof, Ethereum can indirectly verify the validity of all the underlying proofs without having to process them individually, minimizing the verification burden while benefiting from the security and reliability of Aligned Layer.

Moreover, the aggregator component is designed to be flexible and adaptable to different proof systems. It can handle proofs generated by various proving systems and combine them into a consistent format (like Risc0 STARK proof, as mentioned earlier). This flexibility ensures that Aligned Layer can support a wide range of ZK-based applications and use cases, regardless of the specific proving system used.

Recursive Proof Composition

The diagram above illustrates the recursive proof composition process within the Aligned Layer architecture. It shows how multiple proofs are aggregated and verified in a hierarchical manner to create a single, compact proof that can be efficiently verified on the Ethereum blockchain.

• At the bottom of the tree, the initial proofs (Proof 1 to Proof 8) are generated by different proving systems and submitted to the Aligned Layer for verification. Each pair of initial proofs is assigned to a General Prover Verifier, which verifies the validity of the proofs using the corresponding verification algorithms.

• After verification, the General Prover Verifiers aggregate the proofs into a single proof using recursive proof composition techniques. This process continues recursively, with higher-level General Prover Verifiers verifying and aggregating the aggregated proofs until a single, final proof is obtained.

• The Final Proof represents the aggregated and verified proof of all the initial proofs. This compact proof can be efficiently verified on the Ethereum blockchain, reducing the verification burden and cost.

The recursion tree diagram demonstrates the hierarchical nature of the proof aggregation process, where proofs are progressively verified and combined until a single, final proof is obtained. This approach allows for efficient verification of a large number of proofs by leveraging recursive proof composition techniques.

Dual Staking Mechanism

The dual staking mechanism is an integral part of the Aligned Layer architecture, serving two main purposes: bootstrapping the proof-of-stake network and ensuring the long-term security and decentralization of the system.

1. Bootstrapping with Ethereum Staking: Aligned Layer utilizes the EigenLayer framework to enable restaking of Ethereum validators' staked ETH. This allows Aligned Layer to bootstrap its proof-of-stake network securely, attracting validators with additional rewards while benefiting from Ethereum's security.

2. Native Token Staking for Governance and Sovereignty: Aligned Layer introduces a native token for governance and long-term sovereignty. Token holders can participate in decision-making processes, such as protocol upgrades and parameter changes. Native token staking also provides an additional layer of security and incentivizes long-term commitment to the network.

The dual staking model combines the advantages of Ethereum's established staking infrastructure with the benefits of a native token for governance. This approach enables Aligned Layer to bootstrap its network securely while transitioning towards a more decentralized governance model.

By requiring a significant stake in the native token, the protocol creates a strong deterrent against malicious behaviors, as attackers would need to acquire a substantial amount of tokens to potentially compromise the network's security or liveness.

In terms of cost, both soft finality and hard finality modes in Aligned Layer are cheaper compared to Ethereum alone. This cost efficiency is a significant advantage, as it makes ZK proof verification more accessible and cost-effective for projects with limited resources. By reducing the cost barrier, Aligned Layer encourages wider adoption and experimentation with ZK technologies.

The security aspect is crucial in the comparison between Aligned Layer and Ethereum. While Ethereum provides the base level of security, Aligned Layer with soft finality offers security proportional to Ethereum. This means that the security of soft finality mode is derived from the underlying security of Ethereum, providing a strong foundation for trust and reliability. On the other hand, hard finality mode in Aligned Layer inherits the full security guarantees of Ethereum, ensuring the highest level of assurance for the verified proofs.

Moreover, the flexibility to choose between soft finality and hard finality modes allows projects to optimize their verification process based on their specific needs. Applications that prioritize speed and cost-effectiveness can opt for soft finality, while those that require the highest level of security can choose hard finality mode. This opens up more and more use cases for Aligned Layer to explore and to be relied on.

Supported Proof Systems

Aligned Layer's support for multiple proof systems is a significant advantage, as it allows developers to choose the most suitable system based on their specific requirements. Whether a project prioritizes proof size, verification time, post-quantum security, or prover efficiency, Aligned Layer provides the flexibility to accommodate different needs. This versatility makes Aligned Layer an attractive platform for a wide range of ZK applications, fostering innovation and adoption in the ZK ecosystem.

In addition to the proofs mentioned in the table, Aligned Layer quickly adopted the Jolt proof and SP1 into its operations.

Now that we understand Aligned Layer Architecture, the basic flow of proof verification in Aligned Layer can be summarized as follows:

1. Proof Submission: Clients submit proofs generated using various proving systems to Aligned Layer.

2. Data Availability: Submitted proofs are stored securely in the DA Layer.

3. Operator Assignment: Proofs are assigned to operators for verification using a fair allocation mechanism.

4. Retrieval and Verification: Operators retrieve proofs from the DA Layer and run specific verification algorithms to validate the proofs.

5. Consensus: Operators engage in a consensus mechanism to ensure the majority agrees on the validity of each proof.

6. Posting Results: Verification results are posted to an Ethereum smart contract for transparency and immutability.

7. Aggregation: Verified proofs can be aggregated using recursive proof composition techniques, reducing verification costs on Ethereum.

Current Stats around Aligned Layer

Over the past few weeks, the Aligned Layer team has answered several questions regarding their operations, testnet, and partnerships. We have summarized it all here.

Concerns around how the gas savings of verifying proofs on Aligned Layer compare to verifying them on a layer 2 solution with cheaper gas costs.
Based on their current benchmarks, Aligned Layer aims to achieve a 90% reduction in costs compared to verifying proofs on an L2. The computation will be performed in a dedicated network specifically designed to run verifiers, which take around 100 milliseconds on a normal computer. The primary cost of using Aligned Layer will be the data posted on Ethereum, including the hash of the proof commitment, the public inputs, and the true/false result. There are also challenges in verifying BLS signatures, which is a common problem faced by many zkEVM-based applications but this issue has been fixed in zkEVM, and Aligned Layer has solutions in place to scale the verification process with the number of operators.

What are the incentives for using different proof types and the fee mechanism in Aligned Layer?
It should be noted that within Aligned Layer, all proof types are treated the same. The fees for verification will be based on the cost of verifying the most expensive proof type, which currently is STARK proofs. Users will need to pay the fees in advance. The Aligned Layer system is highly scalable, so there is no reason to anticipate long-term problems with fees. The system is designed to handle a large number of proofs efficiently, ensuring that the fee mechanism remains sustainable.

Slashing Conditions in Aligned Layer
Aligned Layer employs a slashing mechanism to discourage malicious behavior and maintain the integrity of the verification process. The slashing conditions are designed to penalize operators who act against the protocol's rules or provide incorrect verification results. There are two main approaches to slashing in Aligned Layer:

1. (Short-term Plan) Subjective Slashing: In the short term, when Aligned Layer goes into production (still in testnet), a simple sub-slashing mechanism will be implemented. It relies on the consensus of a supermajority (e.g., two-thirds) of the total operators to finalize and post the verification results to Ethereum. Under this mechanism, if an operator is in the minority and reports proof as incorrect while the majority of operators deem it valid, the minority operator will get slashed. It should be noted that, in this instance, this sub-slashing mechanism is not perfect because it doesn't leverage all the features provided by zkEVM. However, it serves as a starting point to ensure the security and integrity of the verification process.

The subjective slashing mechanism works as follows:

• Operators participate in the verification process and submit their verification results.

• A supermajority consensus is reached among the operators on the verification results.

• Operators who provide verification results that differ from the supermajority are identified as potential malicious actors.

• The protocol slashes the stake of the non-compliant operators, penalizing them for their malicious behavior.

While subjective slashing is not perfect, as it doesn't leverage all the features provided by zkEVM, it serves as a starting point to ensure the security and integrity of the verification process. The effectiveness of subjective slashing relies on the assumption that the majority of operators are honest and act in the best interest of the network. As the Aligned Layer client side is designed to be lightweight with low hardware requirements, it encourages a more decentralized network with a larger number of operators, reducing the risk of collusion.

2. (Long-term Plan) Objective Slashing: In the long term, Aligned Layer aims to implement a more advanced slashing mechanism. Objective slashing involves running the verification computation directly on-chain using zkEVM. In this approach, the slashing conditions are defined based on the specific proving systems supported by Aligned Layer. Since not all proving systems integrated with Aligned Layer can be fully verified on-chain in Ethereum, objective slashing may not be feasible for all proof systems. However, for proof systems that can be verified on-chain, such as Cairo proofs, objective slashing can be implemented. The objective slashing mechanism works as follows:

• Operators participate in the verification process and submit their verification results.

• If the majority of operators agree on the verification results, they are considered valid.

• In case the majority of operators act maliciously and provide incorrect verification results, an honest operator or a third party can trigger a slashing event.

• The slashing event initiates a challenge process, where the disputed verification results are fully verified on-chain in Ethereum using zkEVM.

• If the on-chain verification proves that the majority of operators acted maliciously, their stakes are slashed, penalizing them for their misbehavior. 

Objective slashing provides a higher level of security by leveraging the on-chain verification capabilities of Ethereum and zkEVM. It ensures that malicious behavior can be detected and punished, even if the majority of operators collude. 

Each verification will have different on-chain computation costs, so they need to take that into account and test thoroughly to create a robust slashing mechanism.

L2IV is Aligned

We are proud backers of Aligned Layer. As the adoption of ZK proofs continues to grow at an unprecedented pace, there is an increasing need for a robust, efficient, and scalable verification infrastructure. Aligned Layer aims to fill this critical gap by providing a universal and modular verification layer for ZK proofs, enabling developers to build and deploy privacy-preserving and trustless applications with ease.

With its commitment to cost-efficiency, scalability, and security, Aligned Layer is set to play a vital role in shaping the future of privacy-preserving and trustless applications, unlocking new use cases and opportunities across the blockchain ecosystem.

About Aligned Layer

• Website: Aligned Layer

• Whitepaper: Aligned

We would like to express our sincere gratitude to Federico, RJ, Francisco, and the entire Aligned Layer team for their invaluable insights into this article

Find L2IV at l2iterative.com and on Twitter @l2iterative

Subscribe now

Author: Arhat Bhagwatkar, Research Analyst, L2IV (@0xArhat)

References

• History of zero knowledge proofs

• Proof aggregation techniques

• An overview of the Groth 16 proof system

• All you wanted to know about Plonk

• Arithmetization schemes for ZK-SNARKs

• GitHub:

  • EigenLayer implementation

  • Tendermint implementation

  • Compilation of resources from Zero to Aligned

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only and do not constitute an investment recommendation or offer to provide investment advisory services.

Nubit: Addressing Bitcoin's Scalability Challenges

The rapid growth and popularity of Bitcoin inscriptions, NFTs, and other Bitcoin-native applications have led to a significant increase in the amount of data stored directly on the blockchain. This surge in data has resulted in network congestion, rising transaction fees, and scalability concerns. The Bitcoin community has been concerned about these “non-payment transactions” taking up Bitcoin’s network resources that lead to an enormous amount of transactions unsettled in the mempool. Nubit tackles these challenges head-on by offloading the storage burden from the Bitcoin blockchain while leveraging its security and consensus mechanisms.

It happens that such a storage layer is sufficient to address many of the challenges that Bitcoin is encountering today and tomorrow.

• Bitcoin inscriptions and NFTs require nothing but storage. These transactions currently use the Bitcoin script opcode OP_RETURN and the increased storage allowance after the Taproot upgrade in order to store data on Bitcoin and require Bitcoin to do no computation over such data. This can all be delegated to a Bitcoin-native storage layer.

• BitVM, Bitcoin optimistic and ZK layer-2 all require data availability for security. Bitcoin itself can store data but it does not currently satisfy the requirements for such data availability, as it needs to be slashable. A more functional, scalable layer for data availability is needed.

Very importantly, what sets Nubit apart is its Bitcoin-native design, seamlessly integrating with Bitcoin's existing infrastructure and security model. This distinguishes it from the existing standalone or Ethereum-native data availability layers such as Celestia. By being native to the Bitcoin ecosystem, Nubit inherits Bitcoin's robust security properties, such as immutability, censorship resistance, and decentralization. Currently, Bitcoin still provides the best crypto economics security—its market cap is 3x the market cap of Ethereum.

How does Nubit achieve Bitcoin nativeness? One of the key innovations of Nubit is the concept of "Nubit tags," which are small cryptographic commitments stored on the Bitcoin blockchain. These tags serve as references to the actual data stored off-chain on the Nubit network. By utilizing Nubit tags, the system significantly reduces the data size on the Bitcoin blockchain, typically from an average of 300 bytes per inscription to just 30 bytes. Nubit tags can represent up to 8MB of data, which is partitioned into smaller tranches or blocks for storage and aggregation. One transaction can include multiple Nubit tags. This optimization enables more efficient use of Bitcoin's limited block space, allowing for higher transaction throughput and improved scalability.

Nubit's Architecture: Combining Consensus, Data Availability Sampling, and Trustless Bridging

Nubit's architecture is designed to combine Bitcoin-native consensus mechanisms, data availability sampling techniques, and a Bitcoin-native bridge to create a scalable, secure, and efficient blockchain platform that inherits Bitcoin's security properties.

Consensus mechanisms. The consensus mechanisms of Nubit are built to inherit Bitcoin's economic security, tamper resistance, and censorship resistance through two key techniques: Bitcoin Staking and Bitcoin Timestamping. 

• Bitcoin Staking allows Bitcoin owners to directly participate as validators using extractable one-time signatures (EOTS). EOTS ensures accountability by revealing the validator's secret key if they attempt to double-sign blocks at the same height so the validator’s security deposit on the Bitcoin network can be slashed. This mechanism aligns the incentives of validators with the network's security, as they risk losing their staked Bitcoin if they engage in malicious behavior.

• Bitcoin Timestamping creates checkpoints on the Nubit blockchain and records them on the Bitcoin blockchain. This protects against long-range attacks and significantly reduces the unbonding period for staked assets from weeks to less than four hours. By leveraging Bitcoin's immutability and security, Nubit establishes a robust and trustworthy consensus layer. (For more on timestamping, please read: Timestamping)

Bitcoin may have more than one million miners. To support a large validator set and achieve high censorship resistance, Nubit employs a PBFT-based consensus algorithm with SNARK-based signature aggregation. In a network with a large number of validators (e.g., 200,000), passing over signatures between all validators in the P2P network is impractical. Nubit addresses this issue by using SNARK-based signature aggregation, which merges multiple signatures into a single one, similar to what Ethereum beacon chain has used since the Shanghai upgrade. In this approach, validators combine signatures from neighboring validators into a single SNARK proof. By doing so, Nubit reduces the communication complexity to nearly constant, making it feasible for a large number of validators to participate in the consensus. This high level of decentralization and consensus resistance are in alignment with Bitcoin’s vision.

Data availability sampling. Data Availability Sampling (DAS) is integrated into Nubit to scale the network with full storage nodes as well as light clients. As the validator set and block size grow, downloading entire blocks to ensure data availability is infeasible. Nubit adopts the modern DAS techniques deployed in production to address this issue. 

There are a few technical highlights of their techniques. First, Nubit uses Reed-Solomon (RS) codes to encode data, which allows efficient proof of storage. A KZG commitment of the data is also generated to verify the encoded data for integrity.  Second, Nubit uses block dispersion to scale the system further. Now, the leader node of the Nubit network distributes coded chunks to different validator groups, each responsible for storing and managing specific chunks.

Nubit uses two protocols for DAS: 

• Sampling Protocol, executed between a verifier and a data source, involves the verifier requesting random block chunks and verifying them against the KZG commitment in the block header. 

• Decoding Protocol, carried out by a decoder working with the validator group, involves requesting block chunks from validators, verifying their correctness, and reconstructing the full block using RS decoding once a sufficient percentage of chunks is collected.

There will be three types of nodes in the Nubit network that participates in the DAS protocols.

• Validators, who run the sampling protocol within the validator set and only sign the block header if it can verify that the data is indeed stored; 

• Full Storage Nodes, which store the entire block and respond to chunk requests from light clients; and 

• Light Clients, which obtain block headers from validators, perform the sampling protocol with full storage nodes, and consider the block available if the protocol succeeds after sufficient sampling.

Bitcoin-native bridge. Nubit implements a trustless bridge using a Payment Channel Network (PCN) based on the Lightning Network. This approach reduces communication overhead, simplifies deployment, enhances security, and provides an emergency withdrawal mechanism for users. The trustless bridge utilizes a hashed-timelock-contract (HTLC)-based protocol using Verifiable Delay Functions (VDFs).

This bridge involves two main components: 

• Fee Payment, where users probe a payment path, assemble time-lock contracts with VDF challenges, and update channel states once the VDF evaluation result is revealed, and 

• Distributing Validator Rewards, where the validator (payee) performs the routing and sends the time-lock contract for the channel connecting the base address and its adjacent node to the consensus nodes for collaborative verification and signing.

The Need for Data Availability Solutions is Real

The growing adoption of data availability (DA) layers can be attributed to the pressing need for scalability solutions in the overall growing on-chain ecosystem. As the adoption of decentralized applications (dApps) and the demand for blockchain-based solutions continue to grow, the need for efficient and scalable DA layers has become increasingly apparent. Also, as Ethereum continues to experience high transaction costs and limited throughput, developers are seeking alternatives that can alleviate these pain points without compromising security or decentralization. Layer 2 scaling solutions, such as rollups and sidechains, have emerged as popular approaches to address Ethereum's scalability issues. However, these solutions often rely on separate consensus mechanisms and require additional trust assumptions.

In this context, Nubit differentiates itself from other data availability solutions in several key aspects. 

In the data availability layer space, Celestia and Avail have gained significant traction. Celestia provides a modular data availability layer for various blockchains and layer 2 solutions, utilizing PoS consensus and DAS. Avail, on the other hand, leverages KZG commitments and polynomial commitments to enable efficient verification and reduce storage burden on nodes. (Read more here: DA Layers)

Nubit’s unique competitive strength lies in its Bitcoin nativeness. This is why it does not use the consensus protocol that Celestia, EigenDA, Avail, and NearDA uses, but it adopts a solution that can scale to a large number of nodes. This is, in essence, closer to Ethereum, which tries to support at least hundreds of thousands of validator nodes. However, even this posed a challenge to Ethereum beacon chain, as verifying a large number of signatures for the entire network would lead to slower finality. This is why one of our portfolio companies, Polyhedra, has been working on single-slot finality for Ethereum using SNARK. We think Nubit is likely going to front-run Polyhedra on this, through their SNARK-based signature aggregation solution. 

This choice of consensus mechanism is crucial for Bitcoin alignment for trust minimization and the need for fast finality, which solves the problem of slower finality (10 minutes) of the Bitcoin network for Bitcoin inscriptions, BRC-20, and similar applications whose primary goal is data storage.

Our Conviction

An important reason that drives our decision to invest in Nubit is the team's track record, expertise, and background. The co-founders, Yu Feng, Hanzhi Liu, and Hongbo Wen, bring a wealth of knowledge and experience in computer science, program verification, and blockchain security.

• Yu Feng, the co-founder of Nubit, is an Assistant Professor of Computer Science at UCSB. He holds a Ph.D. in Computer Science from UT Austin. Importantly, he was a co-founder of Veridise, a highly reputable crypto security company in the blockchain industry. His research focuses on developing automated programming techniques to enhance software usability, reliability, and security. Veridise was a good track record for Yu Feng and his other students.

• Hanzhi Liu, the CEO and co-founder, is a Ph.D. student at UCSB advised by Yu Feng. His research focuses on program verification, language semantics, and compiler optimizations. He has authored a significant paper on certifying ZK circuits and has played a key role in large inscription projects on blockchain, the experience of which is directly related to Nubit.

• Hongbo Wen, the CTO and co-founder, is also a Ph.D. student in Professor Yu Feng's Lab at UCSB. He holds a B.S. in Architecture from Tsinghua University and has worked in startups in both AI and Blockchain security.

The academic achievements, industry track record, and deep understanding of the Bitcoin ecosystem exhibited by the Nubit team give us confidence in their ability to execute the vision of Nubit. 

We believe that Nubit has the potential to revolutionize the way data is stored and accessed within the Bitcoin ecosystem. By providing a scalable, secure, and efficient data availability layer, Nubit can unlock new possibilities for Bitcoin-based applications and foster the growth of the Bitcoin ecosystem as a whole.

Our investment in Nubit reflects our confidence in the team, the technology, and the market opportunity. We are excited to support Nubit in its mission to address Bitcoin's scalability challenges and pave the way for a more scalable and accessible Bitcoin network.

As always, we remain committed to backing transformative technologies and exceptional teams that have the potential to drive innovation and shape the future of the decentralized ecosystem. We look forward to working closely with the Nubit team and witnessing the impact they will make in the Bitcoin ecosystem and beyond.

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only, and do not constitute an investment recommendation or offer to provide investment advisory services.

Here’s Our Thesis

The AI industry is experiencing unprecedented growth, expected to reach a market size of $2 trillion by 2030, and market spending on AI training alone is expected to reach $600 billion during the same timeline. As AI becomes increasingly pervasive across industries, the demand for GPU resources to train and fine-tune AI models is skyrocketing. However, the current centralized cloud infrastructure is struggling to keep pace with this demand, leading to supply constraints, high costs, and limited accessibility for many businesses and researchers.

Decentralized compute protocols, like Lumino AI, are emerging as a compelling solution to address these challenges. By leveraging the vast untapped potential of GPU resources distributed across data centers, crypto mining operations, and consumer devices, these protocols can provide a more scalable, cost-effective, and resilient infrastructure for AI workloads.

As AI becomes a foundational technology across industries, the demand for accessible and cost-effective compute resources will only continue to grow. Lumino AI's decentralized approach not only serves this growing demand but also has the potential to unlock new possibilities for AI innovation by enabling a broader range of organizations to participate in model training.

What does Lumino AI do?

Lumino AI acts as a task distribution layer, intelligently breaking down complex model training jobs into discrete subtasks that can be efficiently allocated to a network of Compute Providers (CPs). By aggregating idle GPU resources from a diverse range of CPs, including data centers, public and private clouds, and even consumer-grade gaming rigs, Lumino AI can provide AI developers with cost-effective and readily available compute power.

Its decentralized compute protocol leverages blockchain to facilitate the distribution, execution, and verification of AI training jobs across a network of heterogeneous compute providers (CPs). The protocol breaks down complex model training tasks into discrete subtasks, matches them to eligible CPs based on their attested hardware specifications and software stack compatibility, and ensures the integrity of the results through cryptographic proofs and independent verification.

By aggregating idle GPU resources from various sources, including data centers, public and private clouds, and even consumer-grade gaming rigs, Lumino AI creates a heterogeneous network of CPs. This diversity provides several advantages: Cost-effectiveness, Increased availability, and Flexibility.

How is it better?

Lumino AI addresses several critical challenges faced by businesses and researchers in the current AI compute landscape. 

• Firstly, it offers significant cost savings compared to major cloud providers, with prices up to 90% cheaper for GPU resources. This cost-efficiency is particularly beneficial for startups and small-to-medium businesses (SMBs) that often face higher prices and limited access to GPU instances from centralized cloud platforms.

• Secondly, Lumino AI provides access to a diverse range of GPUs, including advanced options like the Nvidia H100, which are not readily available from major cloud providers due to supply constraints or allocation priorities. By aggregating supply from a network of CPs, including data centers, crypto mining operations, and Filecoin storage providers, Lumino AI can offer a more stable and reliable supply of GPU resources.

From a competitive standpoint, Lumino AI differentiates itself through its specialized focus on AI workloads, cost-effectiveness, and the breadth of GPU options available. While decentralized compute protocols like Gensyn and Akash have emerged, Lumino AI's singular focus on AI training and fine-tuning, combined with its commitment to privacy and security, sets it apart. Compared to centralized cloud providers and AI-specific clouds, Lumino AI offers significant cost savings and access to scarce GPU resources.

The Supply is Justified

The intense competition for GPU resources, particularly Nvidia's high-performance H100 GPUs, further underscores the need for Lumino AI. As major AI labs and tech giants vie for limited GPU supplies, the issue of GPU access has become a hot topic in Silicon Valley. Andrej Karpathy, former director of AI at Tesla and the founding team at OpenAI, recently highlighted the speculation around H100 demand, with estimates suggesting that major players like OpenAI, Inflection, Meta, and cloud providers could collectively require hundreds of thousands of H100 GPUs.

This heightened demand has led to a "GPU shortage" narrative, with cloud providers struggling to keep pace with the rapid adoption of AI technologies. However, Nvidia executive Charlie Boyle clarified that the issue is more of a supply chain problem than a GPU shortage per se, with small components from vendors being the bottleneck rather than the GPUs themselves.

In this context, Lumino AI's decentralized compute protocol offers a compelling solution by tapping into the vast pool of underutilized GPU resources distributed across various CPs. By aggregating supply from data centers, crypto mining operations, and even consumer devices, Lumino AI can help alleviate the pressure on the GPU market and provide a more accessible and cost-effective alternative for businesses and researchers.

Moreover, Lumino AI's partnerships with a diverse range of CPs, including those with access to cutting-edge H100 GPUs, position the platform to meet the growing demand for high-performance AI compute. As the battle for GPU resources intensifies, Lumino AI's ability to provide access to scarce H100 GPUs and optimize their performance through advanced orchestration techniques further differentiates the platform from competitors.

Why we are excited about Lumino AI?

We have great confidence in the team behind Lumino Labs, led by Eshan and Yogesh. They bring a wealth of experience in AI, distributed systems, and entrepreneurship.

Eshan's background in developing advanced AI models at Zesty.ai, combined with his deep involvement with designing cryptoeconomics within the filecoin ecosystem and Yogesh's track record in building scalable data platforms and infrastructure at AWS and Twilio, along with his experience in AI model development, add depth to the team's technical capabilities.

Moreover, Lumino AI's partnerships with a diverse range of CPs, including data centers, public and private clouds, and Filecoin storage providers, position them well to build a robust and scalable network.

Finally, we believe that Lumino AI has the potential to democratize access to AI compute resources and accelerate innovation in the field. By providing a more accessible and affordable infrastructure for AI training and fine-tuning, Lumino AI can empower a wider range of businesses, researchers, and developers to experiment with and deploy cutting-edge AI applications. Additionally, their focus on user experience and developer enablement through SDKs and intuitive interfaces will be key to driving adoption and overcoming potential barriers to entry.

Our investment in Lumino AI reflects our strong belief in the transformative potential of the emerging DePIN landscape within AI. We are excited to partner with the talented team at Lumino AI and support their mission to revolutionize AI compute.

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only, and do not constitute an investment recommendation or offer to provide investment advisory services.

Here's why we invested in Rome Protocol

Rome Protocol is a shared sequencer infrastructure that leverages Solana to enable cross-rollup liquidity and interoperability for Ethereum L2 scaling solutions.

Currently, rollups operate in silos - each rollup has its own custom sequencer, which orders transactions within that individual rollup chain. This segmentation creates issues like:

1. Liquidity fragmentation: Assets and liquidity pools become isolated within each rollup, unable to move freely across rollups.

2. Complex cross-rollup transactions: It is very hard to execute a single atomic transaction across multiple rollups in a seamless way.

3. Redundant infrastructure: Each rollup team maintaining its own sequencer with validators is inefficient.

Shared Sequencing refers to the process of aggregating and ordering transactions from multiple L2 rollups into a single block, which is then posted to the main blockchain (L1) without executing the transactions.

Rome Protocol aims to solve these problems by providing a “Shared Sequencing” layer that can order transactions across all major rollups concurrently. This is enabled by leveraging Solana as the underlying execution layer (via dedicated Neon EVM smart contract deployments) for the shared sequencer.

Rome Protocol feels existing designs fail hard on two fronts: 

• sequencer centralization and 

• no native interoperability across chains.

Their shared sequencer powered by Solana’s SVM aims higher. Instead of just ordering transactions, Rome's sequencer executes smart contracts to process conditional logic across chains. This unlocks the crypto dream:

• assets seamlessly hopping between rollups, apps splitting work across environments, and 

• atomic composability between fragmented worlds.

Rome's architecture lets the shared sequencer enforce cross-rollup rules and batch transactions atomically across multiple EVMs. So swaps, bridges, arbitrage - you name the complex workflow, and Rome can execute it without hiccups across chains.

The key ingredients enabling this are:

1. A sequencer that executes transactions over just ordering

2. Enforced invariants for atomicity

Rome combines these two masterfully with their Shared Sequencer infrastructure. This crystallizes the vision of deeply integrated, high-throughput rollups.

As major rollups have grown exponentially, the acute pain caused by fragmented liquidity pools has become more evident. When a protocol's liquidity exists fractured across 4-5 rollups, aggregating positions or routing swaps/trades itself becomes highly complex. This is the reason why Arbitrum and Optimism have actively advocated and adopted shared sequencers into their roadmap.

Want Cross-chain Assets to Be Social Butterflies?

Let Rome handle it

Rome's architecture unlocks a cooperative future for rollups 

• Shared Sequencer Input Layer: Responsible for ordering incoming user transactions that are batched within rollups before they go to the next components

• Neon EVMs: Rome Protocol deploys a separate Neon EVM module for each rollup integration. This allows EVM smart contracts execution per rollup state on Solana itself, maximizing efficiency.

• MEV Auctions: This component allows rollup transactions to be ordered and sequenced based on MEV auction mechanisms to optimize for profitability.

• Atomic Composability Layer: This facilitates transactions across rollups that need atomic execution - ensuring they either fully complete or fully fail/rollback. This maintains consistency across chains.

• Transactions are grouped into batches for efficient processing before being submitted to Solana and then to the Ethereum mainchain. Batches are compressed to save gas fees.

Hence, Rome Protocol enables interoperability across rollups by:

1. Using isolated EVMs

2. Aggregating transactions via mempools

3. Sequencing via a unified cross-rollup sequencer

4. Efficient bundled transaction processing on Solana

5. Submitting confirmation back to destination chains

Why Solana?

Solana’s high-speed throughput and parallelization capabilities have been the go-to tech to build on.

1. Cost per Transaction: With Solana's PoH consensus and Sealevel (Solana’s parallel smart contracts runtime), it effectively provides ~$0 cost per transaction, with very high throughput. This is much more efficient than running an EVM per rollup, as other solutions do.

2. Neon EVM allows Ethereum smart contracts to run easily on Solana. Rome Protocol can leverage this ready-made infrastructure rather than having to build their own systems.

3. Solana's industry-leading transaction speeds and throughput capabilities (2000+ TPS currently but poised to keep growing) make it well-equipped to handle the large transaction volumes Rome will need to process across all integrated rollups.

4. With over $1.4 billion worth of SOL tokens staked on Solana, Rome Protocol immediately benefits from the battle-tested security, decentralization, and reorg resistance at Solana's established scale.

L2IV Thoughts

After closely incubating Rome Protocol and observing their impressive progress, we at L2IV are thrilled to formally announce our pre-seed investment. Rome's vision of a shared sequencing infrastructure addressing current rollup limitations compelled us to partner early. 

By elegantly leveraging Solana's capabilities for blazing throughput and highly parallelized transactions, Rome unlocks order-of-magnitude efficiency gains in shared validation, infrastructure, and liquidity across traditionally siloed rollups. Their novel solution centered on a collectively operated (Shared) sequencer executing transactions on behalf of multiple chains profoundly resonated with us. 

Their model of a shared sequencer effectively eliminates the duplication of work that exists due to each rollup producing blocks in an isolated, uncoordinated manner. By consolidating this sequencing task into a shared unit, redundancy is removed.

We are proud to be working alongside Rome as they actualize this reality.

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only, and do not constitute an investment recommendation or offer to provide investment advisory services.

At L2IV, we look for exceptional teams to solve complex problems through innovation. Babylon instantly attracted our attention by pioneering uncharted territory - enabling Bitcoin to provide security and earn yields in proof-of-stake ecosystems.

Here's why we are thrilled to support their vision of a BTC-secured decentralized world and the novel protocols they invent and develop:

The Challenges PoS Chains face today

Babylon’s innovative approach to extracting various facets of Bitcoin security to address key challenges in PoS networks marks a significant advancement in blockchain technology.

1. Addressing Long-Range Attacks

Long-range attacks are a significant vulnerability in PoS networks. In these attacks, malicious validators first unstake from the canonical chain and then create a parallel chain starting from a historical point (often the genesis block) and attempt to present it as the legitimate chain. PoS networks are particularly susceptible due to the low-cost nature of block production, compared to Proof-of-Work (PoW) systems like Bitcoin.

Babylon addresses this risk through a sophisticated checkpointing system anchored in the Bitcoin blockchain. The process involves the following:

• Snapshot Creation: Periodically, Babylon takes snapshots of the state of the PoS chain.

• Data Aggregation and Checkpointing: These snapshots are aggregated and converted into a checkpoint. The checkpoint data is then recorded onto the Bitcoin blockchain.

• Immutable Record: Once recorded on Bitcoin, these checkpoints create an immutable historical record of the PoS chain’s state at specific intervals.

This mechanism achieves two critical outcomes:

• Enhanced Security: The immutable nature of the Bitcoin blockchain means that the recorded checkpoints cannot be altered, making it nearly impossible for attackers to rewrite the PoS chain history.

• Faster Finality: By frequently anchoring the PoS chain state to Bitcoin, Babylon provides a more rapid and robust finality to transactions, bolstering the overall security and integrity of the PoS chain.

2. Addressing Illiquidity of Staked Assets

A key challenge in PoS networks is the illiquidity of staked assets due to the required "unbonding period" – the time it takes for staked tokens to be released back to their owners after being unstaked. This period is crucial for security but results in capital being locked up, reducing liquidity.

Babylon’s integration with Bitcoin helps to address this issue:

• Faster Finality Through Checkpointing: By anchoring checkpoints in the Bitcoin blockchain, Babylon provides a more robust and faster finality for transactions in PoS chains.

• Reduced Unbonding Period: The improved finality and security allow for a shorter unbonding period in the PoS networks that Babylon secures, as the risk of long-range attacks and the need for extended manual social consensus is significantly reduced.

• Increased Liquidity: Shorter unbonding periods mean that assets are not locked up for as long, enhancing liquidity for token holders and validators.

What Stood Apart For Us was Babylon’s Modular Approach to Bitcoin's Structure

Babylon introduces a modular framework to Bitcoin’s traditional monolithic structure by distinctly separating two of its primary functions:

• Timestamping Function: Babylon uses the Bitcoin blockchain as a secure and immutable timestamping service. This function ensures that each transaction or block in a PoS chain can be irrefutably verified against a timestamp recorded on the Bitcoin blockchain. This mechanism provides a historical record and prevents the possibility of rewriting the chain.

• Separately, Babylon leverages Bitcoin as an asset for economic security. This aspect involves using Bitcoin’s market value as a form of collateral or staking in the PoS chains. Validators or participants in the PoS chains can stake Bitcoin as a guarantee of their honest participation in the consensus process.

This modular approach allows Babylon to optimally utilize Bitcoin’s inherent capabilities:

• Data integrity and a synchronized timeline across multiple chains. This approach circumvents the need to engage Bitcoin’s transaction processing capacity for consensus or validation purposes in PoS chains.

• Economic value of Bitcoin harnessed for staking and collateral purposes. This strategy leverages the widespread trust and value of Bitcoin to secure the PoS networks without the need for Bitcoin’s direct participation in their consensus mechanisms.

Babylon’s Bitcoin Staking is a Two-Sided Marketplace

The control plane facilitated by Babylon Chain creates a two-sided marketplace that matches the complementary needs of Bitcoin holders and PoS blockchains:

• Demand Side - PoS Chains PoS ecosystems require robust security to protect integrity and prevent attacks like transaction censorship or rewrite of historical states. By leveraging Bitcoin's superior security properties, PoS chains can fulfill this demand. They pay usage fees in their native token for Babylon's security services.

• Supply Side - Bitcoin Holders Bitcoin holders have an abundance of crypto wealth but limited avenues to productively deploy this for yield generation while retaining exposure to BTC. Babylon's staking protocol unlocks yields for Bitcoin holders by directing their BTC to secure and validate PoS chains as collateral.

Babylon’s control plane thus efficiently matches demand and supply between both sides:

• For PoS chains, it offers security services backed by BTC's hardness without need for asset transfer or derivatives. These include timestamping, fraud proofs, bridges etc.

• For Bitcoin holders, it enables trustlessly deploying BTC to earn yields by validating high quality PoS ecosystems. No third party custodians are involved.

This two-sided model creates a positive flywheel aligning incentives between participants while capturing value.

As more Proof-of-Stake chains integrate Babylon to leverage Bitcoin's hardened security, more fees flow into the platform. These fees fund staking rewards that attract greater amounts of Bitcoin capital into the system as collateral.

In turn, a higher value of aggregated Bitcoin staked means each individual PoS chain enjoys even stronger security guarantees. This leads to higher confidence and further usage by additional PoS networks. This cycle concentrates security where it is most efficient while distributing financial rewards to participants, aligning incentives beautifully across the ecosystem.

Babylon’s Architecture

Babylon's architecture is strategically designed to maximize efficiency, security, and scalability in blockchain operations. This is achieved through a distinct three-layered approach, each serving a specific function within the ecosystem. Babylon's primary technological innovation lies in its unique method of aggregating and transmitting checkpoint data from various Proof-of-Stake (PoS) chains to the Bitcoin network. 

This process involves the following key steps:

• Data Collection (from data plane): Babylon collects block data from multiple PoS chains, including block headers and transaction hashes.

• Aggregation Mechanism (from control plane): This data is aggregated into a single, compact form. This aggregation minimizes the data footprint, addressing the block size limitations of Bitcoin.

• Transmission to Bitcoin (to security plane): The aggregated data is then transmitted to the Bitcoin network, where it is timestamped and stored. This process leverages the robust security and immutability of the Bitcoin blockchain.

Data Plane: The Foundation of Checkpoint Generation

• The Data Plane is the foundational layer of Babylon's architecture, primarily comprising various Proof of Stake (PoS) chains. These chains are responsible for the core activities of blockchain operations, including transaction processing, block generation, and maintaining the blockchain ledger.

• In the context of Babylon's architecture, the crucial function of the Data Plane is to generate checkpoint data. These checkpoints are essentially snapshots of the blockchain state at specific intervals. They provide a comprehensive summary of the blockchain's current status, including validated transactions and the latest block information.

• The PoS chains in the Data Plane are integrated into the Babylon ecosystem. This integration allows for the seamless transmission of checkpoint data from the PoS chains to the Babylon Chain in the Control Plane. The efficacy of this data transfer is vital for the overall functionality and security of the Babylon system.

Control Plane: The Intermediary and Coordinator

• The Control Plane is embodied by the Babylon Chain, a specialized blockchain developed to act as the intermediary in Babylon's architecture. Its primary role is to manage the checkpoint data received from the various PoS chains in the Data Plane.

• The Babylon Chain performs the critical function of batching checkpoint data from multiple PoS chains. This process involves aggregating the data into a manageable format and preparing it for transmission to the Security Plane. The efficiency of this batching process directly impacts the scalability and performance of the Babylon system.

• Once the checkpoint data is batched, the Babylon Chain relays it to the Bitcoin network in the Security Plane. This relay process ensures that the checkpoint data is securely anchored to the Bitcoin blockchain, leveraging its robust security features.

Security Plane: Leveraging Bitcoin's Blockchain for Enhanced Security

• The Security Plane is centered around the Bitcoin blockchain, which serves as the ultimate layer of security in Babylon's architecture. The choice of Bitcoin's blockchain is strategic, given its unparalleled security features, including a vast decentralized network and a high hash rate.

• The primary role of the Security Plane is to ensure the security and integrity of the entire Babylon system. By recording the checkpoint data from the Babylon Chain onto the Bitcoin blockchain, the Security Plane provides an immutable and tamper-proof record of the PoS chains' states.

• The integration of Bitcoin's blockchain in the Security Plane enhances the trustworthiness and reliability of the entire Babylon ecosystem. It ensures that the PoS chains within the Data Plane are protected against various attacks and vulnerabilities, particularly long-range attacks.

This layered architecture not only enhances the operational capabilities of Babylon but also sets a new standard in cross-chain security and scalability. Building on this architecture adapts to new primitive technological innovation.

Babylon's strategic segmentation into distinct planes represents an architectural breakthrough in enabling cross-chain interoperability. By cleanly separating components into modular layers aligned to specific capabilities, Babylon transcends the limitations of monolithic designs.

This principles-based approach reaches the essence of what each constituent does best: the data plane efficiently processes transactions, the control plane specializes in translation and coordination, while the hardened security plane anchors overall integrity.

The elegance is in the simplicity - rather than reinventing redundancies across planes, Babylon conserves resources for specialization.

If you would like to dive deeper into Babylon, here is their whitepaper.

Market Size and Potential

The total staking market cap already exceeds $87 billion for major networks like Ethereum, Aptos, Sui, Cosmos, and $170 billion for the entire PoS networks. This is projected to continue rapid growth in line with crypto adoption trends.

Ethereum alone represents $60 billion PoS market cap with yearly yields nearing $2 billion. As more activity shifts to PoS chains and staking rates rise with improving user experience, total locked value can exponentially multiply.

The recent Glassnode report highlights a significant opportunity for platforms like Babylon to unlock the vast dormant Bitcoin supply to strengthen proof-of-stake security.

The report found that around 66% of the circulating Bitcoin supply, equivalent to 12.26 million BTC, has been completely dormant over the past year without any transactions on-chain. This indicates a staggering amount of BTC that is held long-term but not generating any yield.

While HODLing is often promoted in crypto culture, keeping such vast sums of Bitcoin idle and economically unproductive is inefficient from a broader ecosystem perspective. 

Especially as more activity continues shifting to smart contract platforms and applications built on staking-backed consensus models like PoS, there is a pressing need to reinforce assurances and security guarantees for these ecosystems.

Babylon presents an avenue to redirect some of these dormant Bitcoin holdings to serve a crucial purpose - providing security collateral for the emerging world of PoS-based decentralized apps and protocols where reliability directly impacts user trust and adoption.

Considering over 60% of Bitcoin's supply has been inactive on-chain for an entire year, Babylon taps into an underutilized reservoir that can transform economic security models to power the next generation of crypto innovation.

As the industry matures, BTC staking could emulate wrapped asset growth trajectories. Wrapped Bitcoin's market cap already signifies nearly $6 billion bridged onto Ethereum for DeFi - showcasing user appetite. Babylon can attract similar demand. And with programmable BTC positioned to be a dominant asset across all major smart contract networks, Babylon sits centrally at the intersection of this value migration by channeling staked bitcoin specifically into PoS security.

The Team

Babylon's foundation and development are deeply rooted in academic research and technological expertise, particularly in the realm of Bitcoin security. This is evident from the team's origin, which includes co-founders David Tse, Fisher Yu, Sreeram Kannan (founder of EigenLayer), and other collaborators, all of whom contributed to a research paper on Bitcoin security that laid the groundwork for Babylon.

David Tse (Co-founder)

• He earned a Bachelor’s degree in Systems Design Engineering from the University of Waterloo in 1989, followed by a Master’s degree and a Ph.D. in Electrical Engineering from the Massachusetts Institute of Technology (MIT) in 1991 and 1994, respectively.

• David’s professional tenure at Stanford University is distinguished by his focus on information theory and its diverse applications. His contributions to the field have been recognized with prestigious awards and honors. 

• Notably, he received the Claude E. Shannon Award in 2017, a highly esteemed accolade in the field of information theory. Furthermore, in 2018, he was elected as a member of the National Academy of Engineering, underscoring his significant contributions to engineering and related fields.

Mingchao (Fisher) Yu (Founding CTO)

• Fisher Yu holds a Ph.D. from The Australian National University, where his research centered on network information theory and coding.

• His specialization in the development of theory and algorithms is particularly focused on wireless communication, indicating a strong background in complex network systems and communication technologies.

We would like to extend our sincere gratitude to David Tse, Fisher Yu, and the Babylon team for providing their invaluable insights and cooperating throughout our evaluation of Babylon.

We couldn't be more thrilled to partner with Babylon on their journey of bridging the gap between Bitcoin and the extended PoS ecosystem. We foresee immense potential as this unlocks previously siloed liquidity and fortifies assurances for the next primitives of decentralized infrastructure.

We look forward to entering this promising journey together with David and Fisher as partners.

Disclaimer: This content is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisors as to those matters. References to any securities or digital assets are for illustrative purposes only, and do not constitute an investment recommendation or offer to provide investment advisory services.