PineDev
Log inGet started
For Claude Pro & Max · TradingView Pine Script v6

Validate your strategy like a quant.

Connect your Claude, describe your idea, and get an honest, professional verdict on whether it works — before your money finds out.

Connect ClaudeHow it works

Opening-range breakout

ES1! · 15m · 5 trials counted

Validated

Backtest PF

1.46

Deflated Sharpe

98.7%

Out-of-sample PF

1.37

Monte Carlo band, in-sample trades

  • Enough trades
  • In-sample profit factor
  • Deflated Sharpe ratio
  • Monte Carlo drawdown (95th pct.)
  • Survives doubled costs
  • Not carried by a few trades
  • Consistent across periods
  • Out-of-sample profit factor
  • Out-of-sample efficiency
  • Out-of-sample drawdown in range

RSI + MACD + Bollinger combo

BTCUSDT · 1h · 56 trials counted

Rejected

Backtest PF

1.85

Deflated Sharpe

81.2%

Out-of-sample PF

0.47

Monte Carlo band, in-sample trades

  • Enough trades
  • In-sample profit factor
  • Deflated Sharpe ratio
  • Monte Carlo drawdown (95th pct.)
  • Survives doubled costs
  • Not carried by a few trades
  • Consistent across periods
  • Out-of-sample profit factor
  • Out-of-sample efficiency
  • Out-of-sample drawdown in range

Two example verdicts. The one on the right had the better backtest — and failed.

How it works

Three steps. Claude does the work.

  1. 01

    Connect your Claude

    Add PineDev to Claude — desktop app, claude.ai or Claude Code — and log in once. It runs on your own Claude plan.

  2. 02

    Describe your idea

    Claude interviews you, locks your hypothesis before any results exist, and writes the Pine Script v6 strategy for TradingView.

  3. 03

    Get a quant's verdict

    Backtest in TradingView, drop the trade export into Claude. PineDev runs the tests a professional would and tells you honestly if it holds.

What gets tested

The checks a professional quant would run

Pre-registration
Rules and pass criteria locked before the first backtest.
Deflated Sharpe ratio
Is the result bigger than luck, given every variant you tried?
Monte Carlo
How deep the drawdown could plausibly get — not just how deep it got.
Cost & outlier stress
Does it survive double costs, and without its best trades?
One-shot out-of-sample
Tested once on data it has never seen.
Live incubation
Real trades checked against the range the backtest predicts.

Every formula is documented in the methodology.

Pricing

$39/month

  • Unlimited strategy validations
  • Works inside your own Claude — desktop app, claude.ai or Claude Code
  • Pre-registered hypothesis, locked before any results
  • Deflated Sharpe, Monte Carlo drawdown, cost and outlier stress
  • One-shot out-of-sample test and an honest verdict
  • Incubation tracking once you go live

Requires your own Claude Pro or Max plan. Cancel any time.

Questions

Find out before the market tells you.

Connect Claude