We built a backtester honest enough to disappoint us.
Bao Research is a small data-science and engineering team obsessed with one thing: validation over hype. AIM Lab is our flagship — a rules-based backtesting workbench that would rather tell you the truth than tell you what you want to hear.
Make it easy to find out when your edge is fake
The frustration
Bao Research began with a simple annoyance: volatility and AIM tools were either opaque black boxes or spreadsheets held together by hope — and every one of them flattered its owner.
The thesis
A backtester is only useful if it can disappoint you. The most valuable output isn’t a green equity curve — it’s an honest verdict about whether that curve will survive out-of-sample.
AIM Lab
So we built the tool we wanted: five rules-based AIM engines, a 7-way comparison, walk-forward validation, risk gates, and plain-English verdicts that will happily tell you to do nothing.
The principles behind every verdict
Scientific rigor
Hypothesis, test, validate. Every result is benchmarked against buy-and-hold and checked out-of-sample before we believe it.
Engineering excellence
Reproducible runs, adjusted-close pricing, transaction costs included. The plumbing is boring on purpose so the numbers can be trusted.
Radical transparency
No black boxes and no vanity metrics. If a strategy fails a risk gate or looks overfit, the Lab says so — out loud.
Risk-first design
We start from what can go wrong. Guardrails, gates and drawdown are first-class citizens, not footnotes.
5
rules-based AIM engines
7-way
strategy comparison
Out-of-sample
validation on every optimize
Open beta
built by a small team