line bg

Compass Labs partners with Aave for Aave integration into dojo

May 25, 2022

Overview

Compass Labs has been selected to partner with Aave through the Aave grants program to integrate Aave into dojo. Dojo is a python research platform and simulation framework for decentralised finance. Aave integration will enable investors and smart contract developers to access on-chain protocol data and test, train and optimize on-chain strategies and smart contracts.

Aave is a decentralised borrowing and lending protocol. Lenders earn interest by providing liquidity to lending pools, while borrowers can obtain loans by locking their collateral in smart contracts.

Goal

Aave integration into dojo will equip strategists and developers to test and train DeFi strategies and smart contracts in robust real-world scenarios. This means dojo users can make informed decisions on on-chain strategies before capital is deployed and economically stress test smart contract pre-launch. To save users the struggle to access reliable on-chain data, dojo provides on-chain protocol data and facilitates performance monitoring and pool analytics in real-time. Importantly, dojo runs locally, so there is no sharing of strategies or smart contracts with third parties.

Dojo democratizes AI and paves the way for fully automated and highly optimised DeFi such that teams can stay competitive and move fast on stable infrastructure.

We detail key features of the research platform below:

Key features

  • On and off-chain data ingestion: dojo supports life and historic tick-to-tick data for for integrated protocols and implements flexible data loaders.
  • Run at the exact exchange code: identical simulation at the transaction level ensuring protocol adherence. This gold standard approach guarantees that simulations are executed with on-chain protocol logic and mimic real-world scenarios as close as possible.
  • Agent-based learning: modelling of the full ecosystem with adaptive participants and quantitative optimisation through incorporated feedback loop. This means users can optimize strategies and smart contract parameters in a single simulation loop.
  • Market impact models: cutting-edge machine learning to artificially sample transactions to simulate real-world market conditions. This provides understanding on how strategies are affecting wider market conditions and enables economic stress testing of smart contracts pre-launch.
  • Everything runs locally: full experimentation in python on local infrastructure.
  • User metrics dashboard: real-time insights into strategy performance, pool analytics and positions.

The ability to precisely model protocol mechanics, combined with on-chain data access, opens new possibilities for robust real-world scenario testing and performance optimization. Without compromising on security.

Next steps

We’re excited to be contributing to Aave. Stay tuned, as well be sharing more updates soon!

We are hiring!
webflow icon
All Templates