Dark Matter X
Autonomous quantitative research and execution engine with a multi-agent architecture and a hard structural risk layer.
The engineering.
A multi-agent system
Specialized agents handle market-state analysis, routing, risk posture, sizing, and execution.
A structural risk layer
It can reject actions only for enumerated, auditable reasons.
A historical simulation framework
A 10.5-year historical simulation with statistical validation (bootstrap resampling and significance testing) that kills strategies that don’t survive it.
Circuit breakers
Automatic circuit breakers.
The system, end to end.
What it does.
Specialized agents
Market-state analysis, routing, risk posture, sizing, and execution, each handled by its own agent.
Auditable risk layer
Rejections only for enumerated reasons that can be audited.
Statistical validation
Bootstrap resampling and significance testing across a 10.5-year simulation.
Circuit breakers
Automatic stops built into the engine.
Built with.
- Python
- pandas
- NumPy
- Anthropic Claude API
- Provisional patent applications filed on two of its detection and auditing methods.
Dark Matter X is a quantitative research and risk-engineering project. Nothing on this page describes performance, and nothing here is investment advice.
Want something like this? Tell us what you need built.
2 minutes to apply. Personally reviewed within 24 hours.