Intellify

Agentic-DeFi is an advanced decentralized finance (DeFi). We built Agentic-DeFi as a

Intellify

Created At

ETHGlobal New Delhi

Project Description

  1. Backend (Node.js/TypeScript) The backend is a comprehensive API server that serves as the brain of the system: AI Agent System: Multiple specialized AI agents for different DeFi functions: YieldAgent: Optimizes yield farming strategies RiskAgent: Assesses portfolio risks and liquidation threats GovernanceAgent: Manages DAO participation and voting AgenticYieldAgent: Advanced autonomous yield optimization QueryParserService: Converts natural language to structured queries ResponseSynthesizerService: Generates conversational
  2. Frontend (React + Vite) A modern web interface built with: React 18 with TypeScript Tailwind CSS for styling shadcn/ui component library Framer Motion for animations Recharts for data visualization Supabase integration for data persistence
  3. Strategy Layer (Python) Advanced AI/ML components for trading strategies: Web3 AI Trading Agent: Autonomous trading on Uniswap V4 GAN-based Synthetic Data: Generative adversarial networks for market simulation Teacher-Student Distillation: Knowledge transfer from large to small models Reinforcement Learning
  4. Natural Language DeFi Interface Users can ask questions in plain English: "What's the best USDC yield on Polygon vs Ethereum?" AI agents parse intent and route to appropriate services Conversational responses with follow-up suggestions Multi-agent coordination for complex queries
  5. Advanced Trading Strategies Stateless Agents: Independent decision-making for each trade Stateful Agents: Memory-enabled learning from trading history Cross-Chain Intelligence: Automatic network selection for optimal execution Intent-Based Execution: AI understands user goals and finds optimal paths

How it's Made

Architecture Overview We built Agentic-DeFi as a three-tier system that bridges AI intelligence with blockchain execution: Multi-Agent Coordination We implemented three types of agent coordination: Sequential: Agents execute in order (e.g., Risk → Yield → Execution) Parallel: Agents run simultaneously for independent analysis Hierarchical: Master agent coordinates specialized sub-agents

background image mobile

Join the mailing list

Get the latest news and updates