systems and reinforcement
Research and systems for agents that learn and act.
ReForceMind is a small open research and engineering org working on reinforcement learning, agent environments, robotics inference, and ML systems.
We are interested in the full path from the environment an agent learns in to the runtime that eventually has to execute its policy.
| Project | What it is | |
|---|---|---|
runtime |
FlowEdge | C++23 runtime for low-latency robotics policies with fixed memory, deterministic execution, and explicit deadline handling. |
environment |
NetForge RL | Multi-agent RL environment for cyber defense with partial observability, event-driven actions, and reproducible evaluation. |
Current directions include:
- reinforcement learning and multi-agent systems
- agent environments and policy evaluation
- flow-based robot policies and online adaptation
- predictable inference for embodied systems and accelerators
- world models and learned dynamics