An invitation when access opens
Join the list for an early-access invitation. Markets and availability will be confirmed before you choose a plan. There is no fixed launch date yet.
Stop trusting backtests. Start trusting evidence.
Tell the agent your idea. Evrant writes the strategy, backtests it on historical market data and then tries to break it. Every number traces back to a run you can repeat.
- Years of minute-level market data
- Test data sealed until you decide
- Every result reruns exactly
One idea. A record of every test.
Follow an idea through the research workflow. Each step leaves an artefact you can check; the figures below illustrate that process.
- 01Build
From idea to verified code
Describe a strategy in plain English. The agent turns it into a recorded spec and C# that compiles in a sandbox, or you bring your own.
- 02Research
Experiments with receipts
Backtest on years of minute-level market data, then ask why it works. Every experiment ends in a finding that cites the runs behind it.
- 03Validate
Try to break it
Search for stable regions, freeze a configuration and run out-of-sample, walk-forward, Monte Carlo and robustness tests.
- 04Decide
A candidate with a dossier
Filter trades with XGBoost, let the agent run campaigns on a budget, and record your decision against the full evidence.
Why join the waitlist?
Shape what gets built
Tell us what breaks, what is missing and what you would never trust. Questions and feedback go straight to the engineers writing the engine.
Help set the pass criteria
Tell us which failure modes matter in your research. Early feedback helps us choose useful validation defaults.
Before you sign up.
When do I get access?
Signups are coming soon. Join the waitlist and we will email you when your invitation is ready. There is no fixed launch date yet.
How is this different from a normal backtester?
A backtester tells you how a strategy did. Evrant also tries to break it. History is split into development, validation and sealed ranges, every look at test data is counted, and a frozen configuration has to survive out-of-sample, walk-forward, Monte Carlo and robustness tests before it becomes a candidate.
Where do the AI’s numbers come from?
The deterministic engine produces every metric. The agent plans experiments and writes code, and every figure it quotes cites the engine artefact it came from. Anything it cannot cite is flagged. The agent can still misinterpret a result, so review its conclusions. Decisions stay with you.
Do I need to write code?
No. Describe the strategy in plain English and the agent writes C# against the Evrant SDK. You can also write or upload the C# yourself. Both take the same checked build path.
What do I own, and can I use existing code?
You own the ideas, strategies, code and research results you create. You can review the C# and evidence inside Evrant. Existing code must use the Evrant SDK; Python strategies and broker execution are not included. Self-serve downloads of strategies and evidence reports are not available; ask us for an export instead.
Which markets can I research?
The app’s data catalogue shows each market and the dates it covers, with measured gaps. Every market comes with years of minute-level history.
Can I trade a strategy from Evrant?
Not from inside Evrant. It is research software: paper and live trading are not included.
Is Evrant financial advice?
No. Evrant does not recommend trades, and historical and simulated results do not predict future returns. Trading futures carries a substantial risk of loss.
Early access
Bring an idea. Leave with evidence.
Join free for an early-access invitation. No account or payment required. We’ll confirm market coverage before you choose a plan.
You’re on the list. We’ll email you when your invitation is ready.
For your early-access invitation only. Request removal anytime at [email protected]. See our privacy notice.