Getting started with QuantDev.ai
How this site is organized, what to read first, and how to use it if you're serious about building trading systems.
Welcome. This page exists so you don't have to guess where to start.
The short version
QuantDev.ai is an open-source market-microstructure research lab. The work is to turn raw order-flow into predictive market states, test them in public, and package the ones that survive. Two surfaces:
- Research is the lab notebook. Guides, essays, signal walkthroughs, evidence passes, and the methodology, grouped into series you can read in order. Every note names its sample, its label, and its result. Failures are kept.
- Systems is where surviving research becomes a trading system you can paper-trade in your browser before a dollar is involved.
The research is free and always will be. Nothing on this site is financial advice.
If you're new to microstructure
Read the Foundations series in order. It starts with the machinery below the chart and ends with the two habits that keep research honest: knowing why your backtest lies, and knowing what latency actually costs.
Then read Signals. Two well-known order-book signals, walked through carefully enough that you could implement them without hand-waving, including the conditions under which they stop working.
That's a weekend of reading. It will save you months of chart-pattern folklore.
If you already build systems
Skip to Method for how the lab structures a research pass, then Raw atoms for what that method produces when it is actually run: million-event evidence passes, with what locked and what collapsed. Memory is where the surviving atoms become something you can trade on.
Look for the places where the evidence disagrees with what you currently do, and follow those threads.
How the notes are written
Each note carries a type and a level. Guides and essays need no code. Signal notes include a reference implementation. Evidence passes report a specific experiment on a specific sample and are the closest thing here to a lab result. Framework notes describe method.
Where a note has a companion video, the video sits at the top. The note is the complete version; the video is the tour.
One last thing
Trading is a systems problem. The edge is almost never in the formula on the slide. It is in the hundred small decisions around it: how you clean the data, how you price fills, how you resist the urge to tune away a drawdown. Most of what you'll read here is about getting those decisions right.
Next: Market microstructure, explained like you'll build something with it.