ML Infrastructure Engineer, Safeguards

Guides ML infrastructure for safeguards—inference gateways, model serving (GPU/CPU), guardrail and moderation pipelines in the request path, policy enforcement hooks, safety observability, rollout of filter/model versions, and reliability of the protected inference plane. Use when designing or operating safeguard layers on LLM/ML endpoints, deploying classifiers or moderation services, wiring pre/post-filters, scaling safety microservices, or debugging block-rate/latency regressions on the safety path—not for corporate AI policy (ai-risk-governance), building RAG/agents (ai-engineer), adversarial test campaigns (ai-redteam), general CI/CD (devops), app perf profiling (per…