AI hedge fund

Why deploy ai hedge fund on Render?

AI Hedge Fund is a simulation platform that uses multiple AI agents modeled after famous investors (Buffett, Munger, Burry, etc.) to analyze stocks and generate trading signals. It enables educational exploration of AI-driven investment strategies without placing real trades, combining a FastAPI backend, React UI, and PostgreSQL database deployable to Render with one click.

This template deploys a complete AI hedge fund stack—FastAPI backend, React frontend, and managed PostgreSQL—with all service URLs and database connections pre-wired through environment variables, so the 19 investor agents can start analyzing tickers immediately. Instead of manually provisioning three services, configuring CORS, and connecting a database, Render's Blueprint handles it in one click with DATABASE_URL, VITE_API_URL, and FRONTEND_URL automatically injected across services. You just supply your LLM and financial data API keys; Render handles the managed Postgres, CDN-hosted frontend, and keeps everything grouped in a single dashboard project.

Architecture

What you can build

After deploying, you'll have a web app where you can input stock tickers and get AI-generated trading signals from 19 agents modeled after well-known investors like Buffett, Munger, and Burry. Each agent analyzes your chosen stocks and contributes to a final portfolio recommendation, with all runs persisted to a Postgres database. This is a simulation tool for exploring AI-driven investment analysis—it doesn't execute real trades.

Key features

• Multi-agent investor simulation: Runs ~19 specialized AI agents modeled after famous investors (Buffett, Munger, Burry, etc.) that analyze tickers and produce trading signals with a final portfolio decision.
• One-click Blueprint deployment: A single render.yaml provisions FastAPI backend, React frontend, and managed PostgreSQL with all service URLs and DATABASE_URL auto-wired between components.
• Multi-LLM provider support: Supports OpenAI, Anthropic, Groq, Google Gemini, DeepSeek, OpenRouter, xAI, Kimi, and GigaChat with automatic model filtering based on configured API keys.
• Auto-wired CORS configuration: FRONTEND_URL is automatically injected from the static site's URL, locking down the backend to allow only that origin plus localhost for development.
• Financial data integration: Connects to Financial Datasets API for real market data (stock prices and fundamentals) to feed the AI agents' analysis.

Use cases

• Finance student simulates multi-agent trading strategies for a class project
• Developer explores LLM orchestration by deploying famous-investor AI agents
• Hobbyist backtests portfolio ideas using AI-generated trading signals
• Startup founder demos AI investment analysis to potential investors

What's included

Prerequisites

• LLM API Key: Your API key for the AI language model provider (OpenAI, Anthropic, Groq, Google Gemini, or DeepSeek) that powers the hedge fund agents.
• LLM Provider: The name of your LLM provider (e.g., OpenAI, Anthropic, Groq, Google, DeepSeek); leave blank if using OpenAI.
• Financial Datasets API Key: Your API key for Financial Datasets to fetch real market data including stock prices and fundamentals.

Next steps

1. Open the ai-hedge-fund-web URL and run an analysis with 2-3 tickers (e.g., AAPL, MSFT) — You should see each investor agent (Buffett, Munger, etc.) return trading signals and a final portfolio recommendation within 30-60 seconds
2. Test the backtester by selecting a completed run and clicking backtest — You should see historical performance metrics and a chart showing how the AI signals would have performed
3. Configure a different LLM provider in the ai-hedge-fund-api Environment tab and re-run the same tickers — You should see the model picker reflect the new provider and results may vary based on the different AI model