Launched this week
Reamer Research
Deterministic research engine for mid-frequency quants
4 followers
Deterministic research engine for mid-frequency quants
4 followers
A research engine for mid-frequency strategies on OHLCV bars. Test, diagnose and sweep in Python or C++, then carry the logic live with Reamer Server. A fixed seed gives byte-identical results, slippage included. Linux and macOS. $1,800/yr; 30-day trial $225.
I built Reamer Research because strategy results should reproduce exactly. Run the same strategy with the same seed and you get byte-identical output, stochastic slippage and spread included. It links into your own process as a C library, and you write strategies in Python or C++: a working Python binding and a C++ wrapper ship in the kit. I measured it on a 64-core AMD EPYC: 2,000 runs, 0.59% throughput variation, 20 of 20 runs byte-identical by SHA-256.
The method is in a public whitepaper (doi.org/10.6084/m9.figshare.33972466), and every raw output is archived (doi.org/10.6084/m9.figshare.33877588).
Built for mid-frequency quants who write code and run strategies on bars, intraday to multi-day.