Rust ML LLM Debugging

Debug ML/LLM behavior by separating provider, prompt, tool, retrieval, model runtime, validation, budget, observability, and deployment layers.

Process

1. State expected behavior, actual behavior, impact, and the smallest reproducible symptom.
2. Gather evidence from code, logs, traces, metrics, generated output, database plans, config, or failing tests before proposing a fix.
3. Isolate the boundary that fails and compare it with a known working path.
4. Implement one root-cause fix with focused verification or a regression test where feasible.

Coordination

Use this skill together with itsol-task-intake for ambiguous work, itsol-self-review before handoff, and focused security-* or infra-* skills when the change touches trust boundaries or deployment behavior.

Focused References