krasserm

Run a team of coding-agent sessions toward one goal - spawn workers and sub-coordinators via asn, brief them, monitor by observation, synthesize results. Use only when explicitly asked to form or run a team of agent sessions, or when acting as the coordinator…

Manage coding-agent sessions (Claude Code, codex, ...) programmatically via the asn CLI and tmux. Use when starting, monitoring, or sending messages to agent sessions, or when delegating tasks to parallel agent instances.

Autonomously research, write, and ship ML code on the Hugging Face ecosystem: training, fine-tuning (SFT/DPO/GRPO/LoRA), data processing, inference, and evaluation. Use whenever the task involves training or fine-tuning a model, building or auditing an HF…

Autonomous, budget-bounded loop that runs MANY ML experiments toward a goal with no human in the loop — hypothesize, train on HF Jobs, evaluate, keep/ discard, repeat until the target metric or budget is hit. Requires a `program.md` task spec (goal,…