Intro

Welcome to Finny

Finny is an AI trading assistant. You describe a trading idea in plain English; Finny generates Python strategy code, runs backtests, and helps you iterate — all from your terminal.

Finny is built for traders and developers who want to move from idea to inspectable algorithm without abandoning the terminal. Generated code stays open and reviewable, backtests run against real historical data, and broker connections — when you choose to add them — sit behind explicit configuration on your machine.

What you can do with Finny

Generate strategies

Describe an idea in natural language. Finny writes complete Python strategies into a versioned algo workspace.

Institutional backtests

engine_v2 produces 60+ metrics across return, risk, drawdown, exposure, and stability — plus walk-forward, sweep, and Monte Carlo.

Autonomous research

Research subagents fetch OHLCV data, classify the market regime, and pull news from web search and Discord before the builder runs.

Scheduled watchers

Cron-style subagents monitor deployed strategies with a material-change gate so you only hear from them when something actually moved.

Native brokerage support

Link Alpaca, Binance, or IBKR for paper or live trading. The picker auto-switches on asset-class mismatch.

Bring your own key

Use Claude, OpenAI, Gemini, or a local model. Configure the provider during setup or with /init.

Quick start

Install with the one-line curl installer (no Node required):

bash

curl -fsSL https://finnyai.tech/cli/install | bash

Then launch the interactive CLI:

bash

finny

On first run, Finny captures an email for release notes, lets you pick a path, and starts a welcome session. Connect an AI provider when prompted. Full guide: Install.

Where to go next

Install — prerequisites and first run
Agent Modes — Build, Research, and Chat
Algo Workspaces — per-algo storage, versioning, and logs
Strategy Contract — Shape-C and the broker API
Research Subagents — data extractor, news, regime briefs
Scheduled Watchers — cron monitoring with a material-change gate
Commands — the full slash-command reference
Backtesting — engine_v2, walk-forward, sweep, Monte Carlo
Brokerages — Alpaca, Binance, IBKR
Portfolio Builder — the hidden robo-advisor mode
Plans and Licenses — Free, Plus, Pro, Enterprise, and finny-pro activation

Product positioning

Finny is built for traders and developers searching for AI backtesting, trading strategy generation, and terminal-native market automation. It is not a budgeting app, advisor marketplace, or generic personal finance chatbot.

Relevant search intent

AI trading assistantAI backtesting toolalgorithmic trading AItrading strategy generatorPython trading strategiesautomated market analysis

Differentiation

Not advisor prospecting: Finny targets strategy builders, not wealth-management lead generation.
Not budgeting: Finny focuses on market workflows, generated Python, and backtesting — not expense tracking.
Technical depth: docs, commands, templates, broker details, and backtesting metrics support topical authority.
Transparent workflow: generated strategy code stays inspectable before any broker-connected execution.