GenesisHF

GenesisHF

AI agent run Hedge Fund built on Arbitrum Stylus.

Created on 14th February 2025

GenesisHF

GenesisHF

AI agent run Hedge Fund built on Arbitrum Stylus.

The problem GenesisHF solves

Our project tackles a real pain point in DeFi - the complexity and fragmentation of investment strategies. We built a smart contract-based hedge fund that makes it dead simple for users to get exposure to a diversified crypto portfolio. Instead of manually juggling different tokens and protocols, users just deposit ETH and get shares that represent a professionally managed portfolio. The contract automatically splits their investment across different risk levels - from conservative lending strategies to more aggressive token positions. Think of it as a set it and forget it approach to DeFi investing, but with the transparency and control that comes with being on-chain.

Challenges I ran into

The journey wasn't exactly smooth sailing. Here are the main hurdles we faced:
Gas Limits on Mainnet: The biggest headache came when trying to deploy to mainnet. Stylus is still under development, and we hit some gas limit issues that weren't apparent during local testing. We had to pivot to testnet deployment to keep development moving forward. The issue was with the amount of gas required being substantially higher than the required for deployment, $150 worth of ETH compared to $1.5.
Complex State Management: Building a system that handles user shares, multiple asset types, and maintains the right allocation ratios was tricky. We had to be super careful with the math to ensure users always get their fair share during withdrawals.
Protocol Integration: Since we were working on testnet, we couldn't just plug into existing DeFi protocols. We had to create our own test versions of swap pools and lending protocols to simulate the full functionality. This meant extra development work to create these mock contracts.
Risk Management: Designing the allocation strategy wasn't just about writing code - it required careful thought about risk profiles and how to balance potential returns with stability. We ended up using AI agents to help generate optimal distributions, which was an interesting solution to a complex problem.

  1. Testing Edge Cases: The rebalancing function was particularly challenging to get right. We had to account for scenarios like partial fills, failed swaps, and making sure the contract always maintains the target allocation ratios even when dealing with leftover ETH.
    These challenges pushed us to think creatively and ultimately led to a more robust solution, even if it meant making some compromises along the way.

Tracks Applied (6)

DeFAI: Best-In-Class

Our DeFAI project works end to end, does everything accurately and has insane potential to be an actual working product....Read More

DeFAI: Best DeFAI agent on Arbitrum

Our project works end to end, and is deployed and presented on the arbitrum sepolia network, utlilizes arbitrum stylus c...Read More

Treasury Management: Best Overall

our project does treasury management end to end, and it looks cool and uses rust btw ;) So we think we deserve to be in ...Read More

AI Advancement: Rust/Stylus Agent

Our whole project is based on arbitrum stylus. And it works! Hence we think we deserve a spot here :)

DeFAI: Autonomous Business Agents

Our project works end to end and does treasury management for users. We think this can be a legitimate product hence it ...Read More

Treasury Management: Advanced Portfolio Management

our project does treasury management end to end, and it looks cool and uses rust btw ;) So we think we deserve to be in ...Read More

Discussion

Builders also viewed

See more projects on Devfolio