LLMOps Basic — 135: codify rollback triggers on `Human-in-the-loop handoffs` — memo `680371 [135]`

Lesson 135: Human-in-the-loop handoffs

Focus

Document interfaces between retrieval, prompts, and policy engines. Token Human-in-the-loop handoffs:135 keeps neighbouring lessons differentiable.

Key ideas

• Thread: Human-in-the-loop handoffs · drill v5 · spin 226894.
• 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

Eval harness snippet

case_id: LO-23368
route: support_rag_v5
must_include_patterns:
  - "\[chunk_"
forbid_patterns:
  - "guaranteed SLA"
judge_profile: tempered_4

Practice

Practice Simulate degraded retrieval once; capture user-facing fallback copy. — 135 Bump 19.