Reamer Research

0 reviews

Reamer Research is a research engine for systematic, mid-frequency strategies on OHLCV bars, from intraday to multi-day. Test a strategy against years of history, diagnose it trade by trade, sweep it for robustness, and write a full JSON report for every run. Every fill, cost, margin and accounting rule is written down in an execution specification, and a fixed seed gives byte-identical output on every repeat run, slippage and spread included. It is a precompiled library with a stable C interface that runs on your own machine, on Linux x86-64 and macOS on Apple silicon. Worked Python and C++ reference code ships with it, so strategies can be written in Python without writing a binding. On a 64-core EPYC machine a single run processes about 1.7 million bars per second. An uncut recording shows an AI coding agent, working from the kit documents alone, taking a strategy through the full research loop in about 13 minutes. Not built for high-frequency, order-book or options strategies; market data is not included.