software engineering
at agent scale.
The harness and infrastructure to run, coordinate, and observe fleets of coding agents.
vms
early accessthe future of engineering, research and work at scale
A harness orchestrating agents at scale, each agent on its own vm. Dedicated inference capping cost to a minimum.
- 01
the harness
agent |-- agent | |-- agent | `-- agent `-- agent `-- agent
rust core · rlm
- 02
vms
scale · fork · sota perf
- 03
inference
previewgpugpugpuinfinite capacity · 70-80% less than api at saturation