AI-driven quantitative research

Turn a trading idea into a backtested, fee-accurate strategy

Describe an edge in plain language. A team of AI agents codes it, verifies it, backtests it with real exchange fees, and hunts for overfitting — then paper-trades the survivors on live data. You write no code; no capital is ever at risk.

Free plan · no card required · research & paper only
BTC Mean-Reversion · 4H net of fees
Strategy Buy & hold
+38.4%
Net return
1.92
Sharpe
-9.1%
Max DD
61%
Win rate
Backtested on real data from
Binance Bybit Hyperliquid Deribit OKX
The research pipeline

Most ideas should fail. Here, they fail in simulation.

Every idea runs a rigorous gauntlet — real fees, out-of-sample holdout, overfitting checks. The platform is built to reject curve-fits, not celebrate them.

Ideas researched
729
100%
Coded & verified
712
98%
Passed backtest
115
16%
Survived holdout
35
5%
Reached paper
8
1.1%

97% of ideas are abandoned automatically. That discipline is the product — you only ever look at what survived.

What you actually get

A full backtest report, not a signal

Every strategy comes with the analysis a quant desk would run — net-of-fee equity curve, risk metrics, trade-by-trade breakdown, and a return distribution.

BTC Weekly Cash-Secured Put · DeribitHoldout passed
Strategy Buy & hold
Jun 2023 – Jul 2026 · net of Deribit fees & funding
Risk & performance
+147%
Total return
2.31
Sharpe
1.86
Sortino
-12.4%
Max drawdown
68%
Win rate
2.68
Profit factor
Per-trade return distribution — positive skew after fees
Why not do it yourself

The rigor of a quant desk, without the desk

Backtesting your own ideas takes months to do right. Copying a signal group means trusting a track record you can't see. Here's the difference.

Stratmill Do it yourself Signal groups
No coding required
Real fees, funding & slippage modeledIf you build it
Overfitting & holdout checksRarely
Full, verifiable track recordOpaque
Your ideas stay private
Time to first backtestMinutesDays–weeks
From sentence to strategy

Five stages, fully autonomous

01

Describe

Say your edge in plain language. A research agent turns it into a testable, fee-aware hypothesis.

02

Build & verify

A developer agent codes it; four verification layers catch look-ahead bias and broken logic.

03

Backtest

Real maker/taker fees, funding, leverage and liquidation on historical data. Net-of-cost only.

04

Stress-test

Sensitivity, walk-forward, and a one-shot holdout. Overfits are abandoned automatically.

05

Paper-trade

Survivors run live on the same engine as the backtest. No real capital, ever, in v1.

No fantasy fills

Every backtest pays real exchange fees

A strategy that only works at zero cost isn't a strategy. We charge the actual fee schedule of each venue — a round-trip taker costs ~0.10%, so the edge has to clear it.

Venue Instrument Maker Taker Max leverage
BinanceUSD-M Futures0.020%0.050%20×
HyperliquidPerps0.015%0.045%20×
BybitLinear0.020%0.055%20×
BinanceSpot0.100%0.100%
DeribitOptions0.030%0.030%

Funding, slippage and market-impact are modeled too — PnL is always net.

Built for rigor

A quant desk, on autopilot

A team of AI agents

Research, development, QA, analysis, risk and portfolio agents collaborate on every strategy — each an expert in its stage.

Overfitting defense

Four verification layers plus walk-forward and holdout validation. Deflated-Sharpe and cliff-parameter checks reject curve-fits.

Real-market microstructure

Actual fees, funding, leverage, liquidation, slippage and market impact — modeled on real historical order-flow.

Strategy marketplace

Subscribe to proven strategies — full track record visible, source code protected. Publish yours and earn 70%.

Inherited research corpus

Paid workspaces learn from the platform's entire accumulated research — hundreds of tested strategies of hard-won signal.

Backtest–paper parity

Paper trading runs the exact same engine as the backtest, on live data — what you validated is what runs.

Marketplace

Proven edges, not signals

Every listing shows its real backtested track record; the source code stays with the publisher until you subscribe. You get a runnable copy in your own workspace. Share your own strategy and earn 70% of every subscription.

Browse the marketplace
BTC Weekly Cash-Secured Put
Sharpe 2.31 · 68% win · Deribit
+147%
Return
ETH Funding-Carry Neutral
Sharpe 1.74 · 71% win · Binance
+52%
Return
SOL Momentum Breakout
Sharpe 1.38 · 49% win · Hyperliquid
+84%
Return
Pricing

Start free. Scale when it works.

Usage is metered in credits. Free forever to explore; paid plans unlock the platform's research corpus.

Free
$0
  • 500 credits / month
  • 2 active pipeline runs
  • 1 paper-trading slot
  • Learns from your own runs
Get started
Most popular
Pro
$99 / month
  • 8,000 credits / month
  • 8 active pipeline runs
  • 3 paper-trading slots
  • Inherits the platform research corpus
Choose Pro
Max
$399 / month
  • 40,000 credits / month
  • 20 active pipeline runs
  • 5 paper-trading slots
  • Priority research queue
Choose Max
Questions

What you're actually signing up for

Do I need to know how to code or trade?

No. You describe an idea in plain language; the agents handle the quant work. You review results, not code.

Is real money ever at risk?

No. Version 1 is research and paper-trading only — strategies run on live market data in a simulated account. There is no live execution with your capital.

How do you prevent overfitting?

Sensitivity analysis, walk-forward optimization with in-sample/out-of-sample tracking, a one-shot holdout, and deflated-Sharpe checks. Strategies that only look good in-sample are abandoned automatically.

Are the backtests realistic?

Every backtest charges each venue's real maker/taker fees and models funding, leverage, liquidation, slippage and market impact. PnL is always net of costs.

What are credits?

A prepaid unit of platform usage that meters the AI compute your research consumes. Free plans include a monthly allowance; paid plans include more and never-expiring packs are available.

The math behind every verdict

We try to kill your strategy. The ones that survive are worth running.

A good backtest is easy to fake. Every strategy is put through the same statistical tests a professional quant researcher would demand — designed to expose luck, not reward it.

Deflated Sharpe ratio

Catches a Sharpe that only looks good because many variations were tried. Adjusts for the number of trials so lucky fits fail.

Walk-forward optimization

Catches parameters tuned to the past. Fit in-sample, scored out-of-sample across rolling windows — decay shows up immediately.

One-shot holdout

Catches overfitting to the whole dataset. A final untouched slice is tested exactly once — no second chances, no peeking.

Sensitivity & cliff detection

Catches knife-edge parameters. Each input is swept ±20%; if a small change collapses the edge, it’s rejected.

Cost-clearing gate

Catches edges too thin to trade. Average per-trade return must clear real round-trip fees, or the strategy is abandoned.

IS/OOS decay ratio

Catches in-sample flattery. If out-of-sample performance falls too far below in-sample, the result is treated as noise.

Stop guessing. Start testing.

Put your first idea through the pipeline in minutes — free, no card required.

Free plan · no card required · research & paper only

Stratmill is a research and paper-trading tool, not financial advice or a broker. Backtest and paper results are hypothetical and not indicative of future performance. Figures shown are illustrative. Trading involves substantial risk of loss.