For traders who can read Python

You trade hunches. Make them prove it first.

Tell Jo the setup in plain English. Then the gauntlet:

  • Backtest fills at next bar's open, no lookahead.
  • GREEN, CAUTIOUS, or RED. RED means don't trade.
  • Goes live on paper capital, alert on every fill.
  • Real orders via Zerodha: in closed beta.

Free plan: 200K AI tokens · 30 backtest minutes · no card.

backtest.pyNIFTY 50 / NSE2024-07-052026-07-03COMPLETED
THE IDEA, VERBATIM

"Go long the NIFTY 50 when the 20-hour SMA crosses above the 50. Flip short when it crosses back down."

DRAWDOWN · MAX −6.3%
Strategy +23.6%NIFTY 50 (hold) −0.8%
Return+23.55%
Max DD−6.26%
Win rate64%
Trades74
Bars1h
Period24 mo
Signals fill at the next bar's open. NIFTY 50 hourly bars via Zerodha, long and short. Worst month on record: −6.3% (Apr 2026). It's on the chart.

The whole loop lives in one chat.

Idea to live paper trade without leaving the project. No notebook, no spreadsheet, no copy-pasting between five tools.

01

Describe it

Plain English. Entry, exit, universe. Jo asks when something is ambiguous.

02

Jo writes the Python

A real strategy file you can read, edit, and keep. Every version saved, revert is free.

03

Backtest honestly

Next-bar-open fills, costs in bps, intrabar stops. The trade log exports to CSV.

04

Optimize with a verdict

3-phase grid search: train, test, validate. It ends in a call, and red means don't.

GREENCAUTIOUSRED
05

Paper trade live

Real market bars, simulated money. Positions, P&L, and alerts while you sleep.

Backtests that are allowed to say no.

Most backtesting flatters the strategy. Ours is built to disappoint you early, before the market does it with your money.

No lookahead, by construction

A signal on bar N fills at bar N+1's open. Your backtest can't peek at the close it trades on, so live paper results track the simulation.

Stops fill where they'd actually fill

Platform exit rules (stop loss, take profit, trailing stop) execute intrabar against the high and low, on the touch. No end-of-bar mercy.

Validation that can kill the idea

Parameter search runs train / test / validate splits and scores by lower confidence bound. Overfit ideas come back RED, with the reason.

Research beyond the strategy

Run Python cells against your data in a sandbox: rank the Nifty 50 by a signal, chart pair spreads with Plotly, keep every analysis on the research rail.

Your data, connected once.

Authenticate with a provider one time. Jo probes what it can do (historical, live feed) and uses it everywhere: backtests, research, paper trading.

ProviderMarketsBarsAuth
ZerodhaIndiaNSE & BSE equities, futures, optionsDaily + intraday, 1m to 240mOAuth
DatabentoGlobalCME, NYSE, NASDAQ, cryptoTick data, any interval reconstructedAPI key
MassiveUSUS equities & ETFs, fundamentalsDaily, hourly, intradayAPI key

What QuantJo is not.

Every trading tool oversells. Here is our own risk disclosure, kept current.

DISCLOSURE — WHERE WE DRAW THE LINE
  1. Not financial advice. Jo writes and tests code. It does not tell you what to buy. You bring the edge; we make it testable.
  2. Not a broker. We never hold your funds. QuantJo is a technology layer between you and your own data accounts.
  3. No real-money execution yet. Paper trading runs on live market bars with simulated fills. Live execution is in beta behind regulatory clearance, see below.

Shipping next.

The plan is simple: every strategy you validate here should eventually trade real capital here. In the open, dated, and honest about status.

NOW IN BETA

Live execution via Zerodha

Real orders from your paper-validated strategies. In closed beta while SEBI clearance completes.

BETA
IN DEVELOPMENT

Options strategies

NSE and CME options data is already connected. Strategy support lands in the SDK next.

BUILDING
PLANNED

Fundamentals & global markets

Screen on earnings and balance-sheet data alongside price. More exchanges beyond NSE, BSE, and US venues.

PLANNED
PLANNED

Strategy marketplace

Publish a validated strategy with its full tearsheet. Fork someone else's and test it on your own data.

PLANNED

Pricing.

Start free. Upgrade when you outgrow it.

Free
Free forever

Build and backtest any strategy you can describe.

  • 200K AI tokens / month
  • 30 backtest minutes / month
  • All 16 indicators, all providers
Pro
Free during beta

Run a real research workflow.

  • 2M AI tokens / month
  • 300 backtest minutes / month
  • 5 concurrent paper strategies

Full plan details →

Your next idea deserves a verdict.

Under a minute to your first project. Your first backtest tonight.

Free plan: 200K AI tokens · 30 backtest minutes · every feature above.