Lesson 058: Eval harness hardening
Focus
Swap placeholder nouns for your internal service names immediately. Token Eval harness hardening:58 keeps neighbouring lessons differentiable.
Key ideas
- Thread: Eval harness hardening · drill v8 · spin
196775. - Habit: attach a trace_id to every completion you would paste into an ops dashboard.
- Guardrail: add one RACI bullet for prompt or index changes before tomorrow's standup.
Deep dive notebook
Synthetic drill artefacts
Agent choreography
1. Observe bucket `TENANT-5`
2. Max steps `7`
3. Tools: `retrieve, escalate_human, log_decision`
4. Stop: spend cap | keyword `STOP-5`
Practice
Practice Draft three eval assertions QA must pass before prompt promotion. — 58 Bump 11.