← Curriculum track ← Learn hub
Quanta GenAI Curriculum · LLMOps · Advanced

LLMOps Advanced — 038: pair eval regressions on `Agent guardrail tabletops` — memo `672443 [38]`

Lesson 038: Agent guardrail tabletops

Focus

Document interfaces between retrieval, prompts, and policy engines. Token Agent guardrail tabletops:38 keeps neighbouring lessons differentiable.

Key ideas

Deep dive notebook

Synthetic drill artefacts

Agent choreography

1. Observe bucket `TENANT-12`
2. Max steps `7`
3. Tools: `retrieve, escalate_human, log_decision`
4. Stop: spend cap | keyword `STOP-3`

Practice

Practice Paste the worked template into an internal wiki stub and name owners. — 38 Bump 19.