Building an AI quant trading desk that runs locally
Hey, I'm Josh, building AutoQuant. Dropping a note here before launch on June 30 so anyone landing on this page early has the full context.
The short version: AutoQuant is a desktop app where an AI Loop iteratively writes, backtests, and refines trading strategies for you. It runs locally, your data stays on your machine, and you can drop into Python whenever you want.
Why I built it: I've been a senior software engineer for 16+ years and an active options trader (mostly credit spreads, selling premium) for the last few years. Every tool I tried was either a black-box cloud SaaS (Composer, TrendSpider) or a blank IDE that assumed I had a quant team behind me (QuantConnect, Lean). I wanted something in between.
What's there today:
- AI Loop for equity, options, and cross-sectional strategies
- Monte Carlo significance testing so you can tell real edge from overfitting
- Visual Flow Builder for non-coders
- Paper and live trading via Alpaca
A few questions for anyone reading this before launch:
Are there features you'd expect in this category that I might be missing?
What's your biggest frustration with current algo trading tools?
If you currently use Composer, QuantConnect, or similar, what would make you switch?
Thanks for reading. Roast it. Pre-launch is the best time to course-correct.
Replies