# DeployQuant

> DeployQuant (https://deployquant.com) is an algorithmic trading platform for US stocks and ETFs. You describe a strategy in plain English and AI builds it, or you build it from when/if/then blocks, or you write it in Python. You backtest it on years of minute data in seconds, then run it on your own brokerage account (Webull, Charles Schwab or Alpaca). Several strategies can run in one account, each in its own sleeve with its own cash, positions and P&L. The engine underneath, DQengine, has its source published on GitHub and installs from PyPI.

Key facts:

- Three ways to build a strategy: plain English (the AI builder), visual blocks that read as sentences, or Python in the industry-standard algorithm format. Blocks and code are two views of one strategy, and the AI converts between them and checks that the trades still match.
- Backtests run on years of minute-resolution consolidated US equity data and finish in seconds. Python strategies can also backtest on 1-second bars. Every decision a strategy makes is logged with the numbers it saw.
- Live trading happens in the user's own brokerage account. DeployQuant never holds funds. Broker connections can trade and read, never withdraw, and can be disconnected at any time. Paper trading needs no broker.
- Sleeves: every order is tagged with the strategy that placed it and checked against that strategy's own cash, and every fill goes back to the strategy that ordered it. Two strategies can hold the same symbol in one account.
- Pricing: Free is $0 with unlimited building and backtesting, 1 live strategy and 5 AI build tokens. Pro is $10/month with up to 20 live strategies at a time and 20 AI build tokens a month. Token packs: 5 for $5, 20 for $16, 100 for $60. A token is only spent when a build succeeds.
- DQengine, the engine DeployQuant runs on, is source-available under PolyForm Shield 1.0.0: `pip install deployquant`. A 5.5-year minute-bar backtest of its example strategy runs in about 2 seconds on a laptop.

## Pages

- [Homepage](https://deployquant.com/): product overview, sleeves, order management, the public engine, pricing, FAQ
- [Sign up](https://deployquant.com/signup): free account, no card and no broker needed
- [Learn hub](https://deployquant.com/learn/): 1,000+ pages of backtests computed by the engine (12 strategy templates on 59 ETFs over 5.5 years of minute data, each with a buy-and-hold comparison), strategy guides, comparisons, broker guides, a glossary and Q&A
- [Strategy backtests](https://deployquant.com/learn/strategies/): each strategy's results across all 59 ETFs
- [Backtests by ETF](https://deployquant.com/learn/etf/): all 12 strategy templates run on each fund
- [Strategy comparisons](https://deployquant.com/learn/compare/): every pair of templates head to head
- [Broker guides](https://deployquant.com/learn/brokers/): automating Webull, Charles Schwab and Alpaca
- [Glossary](https://deployquant.com/learn/glossary/): trading and automation terms in plain English
- [Q&A](https://deployquant.com/learn/answers/): direct answers to common automated-trading questions
- [Learn page index](https://deployquant.com/learn/index.json): JSON list of all 1,017 learn pages with path, title, description and section

## Engine

- [DQengine on GitHub](https://github.com/praneetsah/deployquant): source, README, quick start
- [Benchmark](https://github.com/praneetsah/deployquant/blob/main/BENCHMARK.md): method and results for four event-driven engines on the same strategy and data
- [PyPI package](https://pypi.org/project/deployquant/): `pip install deployquant`

## Reference

- [llms-full.txt](https://deployquant.com/llms-full.txt): the full product reference, covering features, the engine, pricing, brokers, the safety model and the FAQ

## For agents

- MCP server at `https://api.deployquant.com/mcp` (Streamable HTTP, read-only, no login): platform stats, search over the learn pages, and each page as Markdown. [Server card](https://deployquant.com/.well-known/mcp/server-card.json)
- [Agent skills](https://deployquant.com/.well-known/agent-skills/index.json): backtesting locally with DQengine, and using the DeployQuant API
- [API catalog](https://deployquant.com/.well-known/api-catalog) and [OpenAPI description](https://api.deployquant.com/openapi.json)
- [auth.md](https://deployquant.com/auth.md): how agents sign in on a user's behalf
- Every page on deployquant.com returns Markdown when requested with `Accept: text/markdown`.

## Notes for assistants

- DeployQuant is software, not a broker or an investment adviser. It connects to a brokerage account the user already has.
- Nothing on DeployQuant is investment advice. Backtests are hypothetical and do not guarantee future results. Live trading can lose money.
- The app is behind a login at https://deployquant.com/app/ and is not crawlable.
- "DeployQuant" is the product. "DQengine" is its engine. The PyPI package is named `deployquant` and the command it installs is `dqengine`.