DJayFi

DJayFi is an AI-based DeFi agent that helps users manage their crypto portfolio in a smart and simple way. It is built using multiple agents that work together to take decisions.

First, the Market Agent collects crypto price data. Then the Strategy Agent decides how to divide the investment based on risk level like low, medium, or high. After that, the Risk Agent checks if the portfolio is safe or not. Then the Execution Agent performs the trade using a KeeperHub-style system and shows logs and transaction details.

All the results are stored using 0G decentralized storage, which gives a root hash as proof. This means the data is stored in a decentralized way and can be verified.

The main goal of DJayFi is to show how AI agents can not only suggest but also take actions and store results. It completes the full flow - analyze, decide, execute, and remember.

DJayFi is built using Next.js and Tailwind CSS for the frontend, which creates a clean dashboard UI. Users can select risk level, analyze portfolio, execute trades, and see history.

On the backend, we used Next.js API routes and created different agents like marketAgent, strategyAgent, riskAgent, executionAgent, and memoryAgent. Each agent has its own job, and all agents work together step by step.

We used the 0G TypeScript SDK to store data. Portfolio decisions are converted into JSON and uploaded using MemData. After uploading, we get a root hash which proves the data is stored on 0G.

We also used a KeeperHub-style execution system. It simulates trade execution with retry logic, logs, and gas usage. This shows how reliable execution can be handled.

Overall flow is: User input → AI agents → execution → storage → UI display

Some parts are simulated but structured like real systems, so it can be easily upgraded in future.