Deffi

Deffi is DeFi portfolio manager that makes DeFi accessible to everyone.

Deffi consists of two main agents:

1. Onboarding agent. Which personalizes the user experience based on their risk tolerance and investment goals.
2. Portfolio Manager agent. Which helps the user manage their portfolio and make investment decisions.

Portfolio Manager has its own tools and can help the user itself:

1. Wallet management tools - ask address, balance, send and receive native token, send and receive ERC20 tokens
2. Defi protocols interaction tools - e.g swap, deposit USDC to Morpho vault, etc.

Also he can hire other agents from pre-built set of agents or in future from agents App Store:

Pre-built agents which are currently available:

1. Defi strategies explanation agent - explain how a particular DeFi strategy works, a brief information about protocols involved (TVL, audits, twitter), risks and rewards
2. Defi strategies monitoring agent
   • 📱 Automated daily reports aggregating insights from DeFi-focused Telegram channels and on-chain data
   • 🔔 Configurable alerts for important DeFi events and opportunities

2 and 3 implemented in a companion bot: Companion bot link: Companion bot repo: Test channel

Main app

Repo link:

Technical Stack:

• Coinbase AgentKit for AI agent orchestration and DeFi interactions, customized for using on the frontend. Tools: walletActionProvider, customized Morpho actions provider, Aave actions Provider started
• Coinbase Smart wallet for secure wallet management on the frontend, transactions batching and paying gas in USDC
• Base network (mainnet + Sepolia testnet) as primary blockchain
• LangChain for natural language processing
• React/TypeScript+Vite frontend
• Coinbase offramp for wallet top up using fiat and paying for LLM requests in the future.

Companion bot with DEFI Strategies explanation and monitoring agents:

repo link: how it works

We've been using AWS Bedrock as our core AI infrastructure for it, leveraging several key AWS technologies and integrations:

Core Architecture:

• Amazon Bedrock for LLM orchestration and RAG (Retrieval Augmented Generation)
• Custom Knowledge Base built on AWS to store DeFi protocols data and historical strategies
• AWS Lambda for serverless compute
• DynamoDB for real-time data storage
• Amazon SNS for notifications

Key Technical Components:

1. LLM Integration:

• Used Bedrock's Claude model for strategy analysis and risk assessment
• Implemented custom prompt engineering for DeFi-specific analysis
• Built RAG system to provide context from historical data

1. Data Processing Pipeline:

• Created custom Knowledge Base using Bedrock KB to store:
  • Protocol TVL data
  • Historical strategies
  • Risk metrics
  • Audit reports
• Implemented real-time updates using DynamoDB streams

1. Multi-Language Support:

• Leveraged Bedrock's multilingual capabilities for translation
• Built custom tokenizers for DeFi-specific terminology

1. Integration Layer:

• Built custom connectors for:
  • Telegram API
  • DeFi Llama API
  • Protocol APIs

Notable Technical Achievements:

• Developed a custom RAG implementation that combines real-time DeFi data with historical analysis
• Created an efficient caching system for frequently accessed protocol data
• Built a flexible plugin architecture for easy integration of new protocols

The most "hacky" but effective solution was our custom prompt engineering system that dynamically generates context-aware prompts based on the type of DeFi strategy being analyzed, significantly improving the accuracy of the analysis.