The systematic backtesting platform that doesn't stop at backtesting.
Periscøpe is a systematic backtesting platform for Python strategies. The same code that runs your backtest also runs in paper and live, so a strategy that survives historical data doesn't need to be rewritten to go to production.
What makes a backtesting platform "systematic."
"Systematic" is doing a lot of work in that phrase. On Periscøpe it means four specific things.
Event-driven
Strategies react to market events tick by tick, the same way they will in production. No vectorized shortcuts that paper over what live execution looks like.
Deterministic
Same code, same config, same data window produces the same result. Every backtest is a re-runnable record, not a one-off cell output.
Parameterized
Strategy parameters are declared up front and edited from the run form, not buried in code. Sweep symbols, sizes, and lookbacks without redeploying.
Promotable
The same strategy that ran in backtest runs in paper and live. No second codebase, no rewrite to get to production.
The same strategy code in backtest, paper, and live.
Strategies are written once and promoted across execution modes. Only the data feed and broker change; your strategy code does not.
- Backtest against historical data
- Paper trade against a live feed
- Live trade through a real broker
- One code path, three modes
Every backtest is a first-class run, not a one-off output.
Each backtest is a durable record with config, results, fills, positions, and a classified issue feed — re-openable, comparable, shareable.
- Positions, fills, and P&L for every run
- Orders, ledger, and a classified issue feed
- CSV export for run data
- Re-launchable as paper or live
Backtest → Paper → Live, on one code path.
Backtest
Run the strategy against historical data. Same engine, same events, same actions as live — only the data source changes.
Paper
Wire the strategy to a live data feed but a simulated broker. Watch the system behave in real time, with no capital at risk.
Live
Same strategy, real broker. The promotion changes the feed and the broker; the strategy code does not change.
A Python backtesting platform with AI in the editor.
Strategies are authored in an in-browser editor that understands the strategy SDK and the runtime context, with an AI assistant for reviewing code, explaining failures, and suggesting fixes.
- Strategies in pure Python
- Typed actions and events via the strategy SDK
- AI assistance grounded in the strategy and run context
Where Periscøpe fits.
Backtesting is not the only thing on Periscøpe, and Periscøpe is not the only tool that backtests. Two honest cases:
Backtests you'll keep and runs you'll promote.
Rigorous backtest experiments where each run is a first-class record — config, results, issues — and where the same strategy can move into paper and live without a rewrite. If you want one environment for the whole lifecycle, this is the fit.
Throwaway notebook exploration.
If you're sketching an idea you don't intend to keep, share, or rerun, a notebook with pandas and a backtesting library will be faster. Periscøpe is built for systems you'll operate, not single-cell experiments.
Backtesting on Periscøpe: questions.
What is a systematic backtesting platform? +
A systematic backtesting platform runs a coded trading strategy against historical market data and records what would have happened — every order, fill, position, and P&L. Periscøpe goes a step further: the same code that ran the backtest can run live, so a strategy that survives backtest doesn't need to be rewritten to go to production.
Does Periscøpe support paper and live trading too? +
Yes. Periscøpe runs the same strategy code in three modes: Backtest (historical data), Paper (live feed, simulated broker), and Live (real broker). Promoting a strategy means changing the run mode, not the code.
Which asset classes can I backtest? +
Symbol parameters currently support equities and futures. See the strategy docs for the supported asset types and how to declare symbol parameters in your strategy.
What language are strategies written in? +
Python. Strategies are authored in an in-browser editor with AI-assisted debugging, then published as versioned objects that runs reference.
Can I move from backtest to live without rewriting code? +
That's the design center. Backtest, paper, and live share the same strategy code path. Promoting from one mode to the next changes the data feed and broker bindings, not the strategy.
How is this different from running backtests in a notebook? +
A notebook is great for one-off exploration. Periscøpe is built for runs you keep, share, and rerun: every run is a first-class record with config, fills, positions, ledger, and a classified issue feed. A notebook also can't be promoted to paper or live without a rewrite.
Run your first backtest on Periscøpe.
We onboard in focused cohorts. Tell us what you trade and how you work. Live access is rolled out to select users on request.