AI quant tools FAQ

How to choose, who it is for, and who it is not for. The full order is on how to choose AI quant tools. This page only answers the short questions.

How to choose

How to choose: AI quant work uses four kinds of tools: a data source for prices, a backtesting framework to test a strategy, a trading bot or broker connection to send orders, and a research agent that uses a large language model. Start with one data source and one backtesting framework, get one simple strategy running, then add AI if you still need it. After a backtest, paper trade before real money.

Who it is for: someone who wants one simple strategy running before adding more tools. For personal research on US stocks, yfinance is usually enough. Learning on backtrader is fine. TradingAgents is the original open-source LLM framework and the safest to build on. All of these are research tools.

Who it is not for: someone who wants a profit guarantee, or who wants to skip paper trading and send real orders on day one. Listings, star trends and comparisons are not investment advice, and a backtest is not a promise about future results. None of the LLM agents is evidence that a language model can trade profitably.

Frequently asked questions

What kinds of tools does AI quant work need?

Four kinds: a data source for prices, a backtesting framework to test a strategy, a trading bot or broker connection to send orders, and a research agent that uses a large language model. Decide which job you are doing before you compare products inside that group.

What should a beginner start with?

One data source and one backtesting framework. Get one simple strategy running, then add an AI agent if you still need it. After a backtest, paper trade before real money. You do not need one tool from every category on day one.

Which free market data library should I start with?

For personal research on US stocks, yfinance is usually enough. AKShare is another free library, aimed at public Chinese market pages. OpenBB connects data providers you already have. EODHD APIs is a paid service for when free sources stop being enough. yfinance and AKShare read public websites and can break without notice.

Is backtrader still a good first backtester?

It still installs and still runs, and it is fine for learning. Its repository has been quiet for a long time, so a new project is better on something still maintained: vectorbt for large parameter sweeps, zipline-reloaded for a classic event-driven engine, NautilusTrader when backtest and live trading must share code, or PyBroker for rule and model strategies.

Which open-source LLM trading agent should I look at?

TradingAgents is the original framework and the safest to build on, under Apache-2.0. TradingAgents-CN is a separate A-share project with a web interface; its application layer cannot be used commercially. Vibe-Trading is MIT-licensed, covers more markets, and is the quickest to install. All three are research tools. None is evidence that a language model can trade profitably.

How do I connect an AI agent to an exchange without risking real money?

For OKX, Agent Trade Kit runs on your own computer. Public market data needs no API key. Account data and orders need your own key with read and trade permissions only, and no withdrawal permission. Start with demo keys and read-only mode. The kit places an order only when you or the model asks; it is not a bot that keeps running a strategy.

Can these tools guarantee a profit?

No. Listings, star trends and comparisons are not investment advice, and a backtest is not a promise about future results. Language models can misunderstand an instruction or invent a figure. Treat the output as research, then check it yourself.

Read next

how to choose, in order · a beginner learning path · tools by job · market data libraries · a daily price pipeline · how to choose a backtesting framework · how to choose a quant trading platform · backtrader and maintained alternatives · LLM trading agents · connecting an agent to OKX · portfolio optimization libraries