Observable, deterministic meta-harness for local AI coding agents — supervises Codex and Claude Code through one control plane with static strategy selection, verifier-gated bounded loops, isolated Git worktrees, a durable event trace and the ACR-ARCH selector benchmark.
Read the connected notes for the decisions, experiments and questions behind this project.
Embodied AI is the field: intelligence closing a perception–action loop against physics. Cross-embodiment (X-embodiment) is a scaling bet inside it — and the RT-X numbers say it pays off asymmetrically: 27.3%→75.8% at 55B params, but worse-than-solo at 35M. Diversity beats volume, human video is the cheapest body, and the evaluation harness matters at every scale.
RoboLLM's four principles in practice: one ROS 2 bridge for MCP and dashboard, a physical-arm chain where no model ever touches PWM, gesture pick-and-place and fist-triggered wall welding with collision-verified standoffs — and why it is the same shape as a meta-harness.
Full-body Talos mirror in RoboLLM: pink QP differential IK lost to direct joint copy in an A/B — wrist error 0→5.31 mm, 31–58 ms ticks vs a 20 ms budget, a crash on partial visibility, and an orientation 'win' with no ground truth. Why Pose-only input caps fidelity and when QP actually pays.
AI-assisted money management — planning calendar, Gantt roadmaps, Stripe billing, and generated PDF reports.
Explore projectOn-call incident copilot that drafts timelines and postmortems while responders focus on the fix.
Explore project
