About Wickra Darwin

Wickra Darwin evolves trading strategies with a genetic search — mutating and crossing JSON strategy specs, scoring each candidate with the Wickra backtest engine at millions of backtests per second. An evolution is a JSON document — data, not code — so the search runs in every one of ten languages and, for a given seed, returns a byte-for-byte identical report.

What makes it different

• The genome is data. A population of StrategySpec genomes — the same spec wickra-backtest runs — is evolved with genetic operators (mutation + crossover across the indicator search space). Because it is data, it crosses the C ABI and WASM unchanged.
• Millions of backtests per second. Each candidate is scored by the O(1)-per-tick engine, so the loop sustains a very high backtests-per-second rate — orders of magnitude faster than pandas-based tooling. "AlphaZero for trading strategies."
• The whole indicator space. The search ranges over the 514 indicators of the Wickra core and their parameter ranges — real strategies, not toy signals.
• Seeded and reproducible. A fixed portable PRNG drives mutation and crossover, so a given seed produces the byte-identical report on every platform, pinned by a golden corpus in CI.

Why it exists

Strategy search is usually one language, one bespoke loop, and rarely reproducible. Wickra Darwin defines the search once, in Rust, and exposes it as a JSON-over-C-ABI data API to Rust, Python, Node.js, WASM and — over a C ABI — C, C++, C#, Go, Java and R. The spec is portable JSON, so the same evolution runs anywhere.

Open source

Released under the MIT OR Apache-2.0 license — permissive, OSI-approved, free for any use including commercial. Source, issues and releases on GitHub.

Disclaimer

Wickra Darwin is a software library, not a trading system, and is provided as-is with no warranty. An evolved strategy is a search result over historical data, not financial advice, and past fitness does not predict future returns. Use it at your own risk.