Skip to main content

UniswapV2 Multichain TVL Aggregator Challenge

Context​

Develop a system to aggregate and report Total Value Locked (TVL) from UniswapV2 pools across multiple blockchains. This project aims to create a robust, efficient, and scalable solution for TVL tracking in decentralized finance (DeFi) ecosystems.

Goals​

  • Create a monorepo-based application for multichain TVL aggregation. Just the API, no frontend needed
  • Implement efficient data fetching using virtual batching and multicall techniques
  • Develop a flexible and extensible architecture for future enhancements
  • Ensure high test coverage and deployment readiness
  • Get a grasp on the tech and tools we commonly use

Starting point: Tech design​

Given the starter UML diagram provided, work on a brief tech-design for the challenge

Tech design must address:

  • Which is the plan?
  • 1 Sequence diagram (include diff contracts)
  • 1 UML (relationship between missing modules)

offchain-img.png

Coding fun​

Create the challenge repo using the Wonder's Repo Creator and the Offchain-Turborepo-Boilerplate template. Give it a meaningful name 🙂

Must-do Components​

  • Monorepo Structure:
    • Utilize monorepo template
  • Pricing Module:
    • Fetch prices from Coingecko and DeFiLlama
    • Implement factory pattern for provider selection
    • Add caching with 60-second TTL
  • Logger:
  • EVM Provider:
    • Implement readContract method
    • Utilize Viem as the blockchain interfacing client (we ❤️ viem)
  • Batching System:
    • Develop a batching contract for fetching TVL from different pools of Uniswap using token addresses as inputs
    • Use Foundry for smart contract development
      • Note: don't need to become an expert on SC development
    • Extend EVM Provider with readBatchContract and test your contract implementation
  • TVL Provider:
    • Support multiple chains
    • Implement proxy pattern for caching (TVLProviderWithCache)
  • API:
    • Create a single GET endpoint: /tvl
    • Implement controller and services with dependency injection
    • Return TVL for specified chains based on request body
    • Implement environment variable validation for all apps
    • Swagger docs
  • Testing (this should be done along the way):
    • Use Vitest for testing framework
    • Achieve minimum 90% code coverage
  • Deployment:
    • Provide Dockerfile for local API deployment

Criteria​

  • Code design and modularity
  • Tests, tests and tests!
  • Proper building process
  • Consider error handling
  • Documentation, how can i use each library?
  • Clean project structure and DRY

Next steps​

Write some notes with your experience working on the challenge. All feedback is really welcomed and appreciated, as it help us on improving the challenge for the future