Copilot
Describe a rule in plain English. Copilot writes the Python, runs the backtest in an isolated sandbox, reads the error when it fails and repairs the code.
Research workspace
Quant.
Know before you risk it. Describe a strategy in plain English, test it on market data, and find out whether the result holds up before real capital or evaluation fees are on the line.
From an idea to a verdict
An AI-native workspace for quantitative and systematic traders. It reports the numbers, including the ones that argue against the strategy.
Honest backtests
Orders fill at the next bar's open by default, after commission and slippage. Every result states the fill model and costs it used, flags gaps in the data and compares against a dividend-reinvested buy and hold.
Validation
Bootstrap Monte Carlo, regime analysis and walk-forward re-tests try to break the result, then roll up into an A–F verdict.
Prop-firm simulator
Replay a strategy day by day against one-step and two-step challenge rules, or your own: profit target, daily and total loss, static or trailing, minimum days and a time limit.
Notebooks
Python notebooks for research next to your strategies, running in the same sandbox, so exploration and testing share one set of data.
Other platforms
Syntax checks for Pine Script, NinjaScript, EasyLanguage and PowerLanguage, for strategies written elsewhere. Only Python is executed.
Idea, rules, test, validate, decide
Idea
Rules
Test
Validate
Decide
Quant course
Eleven modules that teach the workflow behind the workspace: from a discretionary setup to Python, honest backtests, validation and a system that runs. Each lesson ends with a Lab exercise and a short self-check.
- 11
- 9
- 2 h 15 min
In private build
The workspace is being prepared for launch. Until it opens, the app button above and the Lab buttons in the course stay disabled; the course itself is open now.