georgezouq/awesome-ai-in-finance

This project is a curated list (an "Awesome" list) of resources for using artificial intelligence in finance and trading. It collects research papers, open-source tools, code, courses, and strategies — all organized into clear categories.

Why it exists

The opening text sets the context: global financial markets generate enormous amounts of data that humans struggle to process. The goal of this list is to gather the best resources that let people use AI to "analyze & trade automatically and intelligently" — and ultimately "beat the market."

What's inside (the main categories)

Agents — Fully or partially autonomous AI trading systems. Examples include multi-agent platforms where several AI agents compete or collaborate, self-improving trading systems, and no-code tools for building trading strategies.

LLMs (Large Language Models) — Resources applying language models to finance: benchmarks that test AI's investing abilities (one gives each model $10,000 of real money to manage), open-source financial models like FinGPT, and research showing GPT-4 outperforming professional analysts in some tasks.

Skills — Reusable code "skills" for AI coding assistants (primarily Claude Code), such as stock research, bias detection in investment analysis, and trading journaling.

MCP Servers — This is a key section. MCP (Model Context Protocol) is a standard way to connect AI assistants like Claude, Cursor, or VS Code Copilot to external data and services. These servers give AI assistants direct access to:

• Market Data — real-time stock prices, SEC filings, economic data
• Trading Execution — letting AI place trades through brokers like Alpaca, Kraken, or Interactive Brokers
• Research & Analysis — tools for technical analysis, earnings calendars, and financial report parsing

Papers — Academic research spanning from classic 1900 work on speculation theory to modern deep reinforcement learning for stock trading.

Courses & Books & Blogs — Educational materials, including university courses (NYU, Udacity), books on financial machine learning and building financial software with generative AI, and practical coding tutorials.

Strategies & Research — Subcategorized by approach: time series prediction, portfolio management, high-frequency trading, event-driven strategies, cryptocurrency strategies, technical analysis, lottery/gambling, and arbitrage.

Other sections — Data sources, research tools, trading system frameworks, technical analysis libraries (TA Lib), exchange APIs, articles, and miscellaneous resources.

Standout entries (marked with stars)

Some listings have star ratings (🌟, 🌟🌟, 🌟🌟🌟) indicating high significance. Notably:

• Nof1 — a benchmark giving each AI model $10,000 of real money in real markets to measure actual investing ability
• MarS — Microsoft's financial market simulation engine
• AI Hedge Fund — a popular project exploring AI-driven trading decisions

Who would use this

Developers, quantitative researchers, and traders interested in applying AI to financial markets. The list is a starting point — a directory of resources rather than a single tool itself.

Important note: The list includes both research-oriented tools (e.g., "no brokerage connection by default") and actual trading systems. Some entries explicitly state they do not execute trades, while others do. The README makes no claims about which approaches actually "beat the market" — it simply curates what exists.