18 terms
Definitions in this topic
- ProductionAgent checkpointAgent checkpoints explained: durable progress records for safe resumption and recovery.
- ProductionAgent circuit breakerAgent circuit breakers explained: automatic failure containment and controlled recovery.
- ProductionAgent kill switchAgent kill switches explained: independent stopping, credential revocation, and recovery.
- ProductionAgent recoveryAgent recovery explained: safe state restoration, retry decisions, and escalation.
- ProductionAgent retry policyAgent retry policies explained: error classes, attempt limits, delay, and safe stopping.
- ProductionAgent rollbackAgent rollback explained: reversing agent changes and handling actions that cannot be undone.
- ProductionAgent timeoutAgent timeouts explained: bounded execution and explicit action after time expires.
- ProductionCompensating actionCompensating actions explained: business reversal when literal rollback is impossible.
- ProductionConfidence gatingConfidence gating explained: why self reported confidence fails, and what signal to gate on instead.
- ProductionDeterministic replayDeterministic replay for AI agents: reconstructing execution from recorded inputs and results.
- ProductionDurable executionDurable execution explained: checkpoints and replay that let long-running AI agents recover without repeating work.
- LLM foundationsHallucinationHallucination as a business risk to gate, not a quirky bug: where it shows up, and how to catch it before a customer does.
- ProductionIdempotent tool callIdempotent tool calls explained: safe retries, stable request keys, and stored outcomes.
- ProductionLLM Output ValidationLLM output validation explained: verify model responses against schemas, business rules, and permissions before production use.
- ProductionLLM Rate LimitingLLM rate limiting explained: request, token, and concurrency limits that protect availability and control inference spend.
- ProductionLLMOpsLLMOps explained: the practices that keep a model or agent reliable in production, and how it extends MLOps.
- ProductionSandboxed code executionSandboxed code execution explained: why agent-written code is untrusted, what the isolation boundary must block, and where the code interpreter fits.
- ProductionTime to first tokenTime to first token explained: what fills the wait before streaming starts, and which levers actually shorten it in production.