ai

Use Your Own AI with Financial Data

Three ways to use your existing AI subscription with Trading Studio: generate SQL queries, fetch data via API, or run and customize 85+ backtesting strategies.

Swas

Swas

— 2 min read

If you have a ChatGPT, Claude, or Gemini subscription, you can use it to work with financial data on Trading Studio. This overview explains the available approaches and which one fits your use case.

What Trading Studio provides

  • Managed data warehouse with FMP financial data (stock prices, fundamentals, analyst estimates, insider trading, ETFs, economic calendars)
  • DuckDB query engine for fast analytical queries
  • API access for programmatic data retrieval
  • Cloud compute for running backtests
  • 85+ open-source backtesting strategies

You can access all of this using your own AI tools instead of (or in addition to) the platform's built-in AI.

Three approaches

Guide 1: Generate SQL with AI (no API key)

Best for: Quick queries, learning SQL, users who prefer web interfaces

How it works:

  1. Copy a compact schema from the Data Explorer
  2. Paste it into ChatGPT/Claude/Gemini
  3. Ask questions in plain English
  4. AI generates DuckDB SQL
  5. Paste the SQL back into the Data Explorer
  6. Run and see results

Setup: None. Just your existing AI subscription.

Use case: "Show me technology stocks with P/E under 15 and ROE over 20%"

Read Guide 1

Guide 2: Query via API (API key required)

Best for: Scripting, automation, repeated queries, Python users

How it works:

  1. Set your API key as an environment variable
  2. Give your AI the endpoint documentation
  3. Ask questions
  4. AI generates and runs Python code that submits queries and fetches results

Setup: API key + Python with requests.

Use case: Building a data analysis script where the AI pulls data programmatically.

Read Guide 2

Guide 3: Backtesting with AI (API key + backtests repo)

Best for: Strategy research, signal testing, parameter optimization

How it works:

  1. Clone the open-source backtests framework (85+ strategies)
  2. Run existing strategies or tell your AI to customize them
  3. AI modifies signal logic, filters, parameters
  4. Run locally or on cloud compute via the Projects API

Setup: API key + Python + backtests repo.

Use case: Testing a custom value strategy on Indian stocks with transaction costs, or running a parameter sweep across 20 configurations on cloud compute.

Read Guide 3

Which guide should you use?

If you're just getting started: Use Guide 1 (SQL generation). No setup, no cost, works with any AI chatbot.

If you're writing a Python script or notebook: Use Guide 2 (API queries). Your AI can generate data-fetching code that you integrate into your analysis.

If you want to test trading strategies: Use Guide 3 (backtesting). Run 85+ strategies as-is, customize with AI, or build new ones from scratch.

You can mix approaches. Many users start with Guide 1 for exploration, then use Guide 2 for scripting once they know which data they need, and Guide 3 when they're ready to test strategies.

API key setup

Guides 2 and 3 require an API key.

Get your key:

  1. Sign in to tradingstudio.finance
  2. Go to Settings, API Keys
  3. Click "Create New Key"
  4. Copy the key (format: ts_xxxxxxxxxxxx)

Set it as an environment variable:

export TS_API_KEY="ts_your_key_here"

Add to ~/.zshrc or ~/.bashrc so it persists across sessions. Then tell your AI: "Use the TS_API_KEY environment variable for authentication."

Resources

Start with Guide 1: Use AI to Query Financial Data

Read more