• Architect and maintain the LLMOps/GenAIOps toolchain, including model registries, prompt version control, and reproducible training pipelines. • Implement and manage the Azure AI Foundry environment, configuring model routers, quota management, and private endpoints for secure inferencing. • Develop comprehensive observability dashboards to track model latency, token costs, hallucination rates, and drift. • Automate "Policy-as-API" controls within the orchestration layer to enforce governance guardrails (e.g., PII filtering) at runtime. • Collaborate with the Platform SRE team to ensure high availability and disaster recovery for mission-critical clinical agents. • Manage the "Model Registry," ensuring all deployed models have associated version history, performance metrics, and rollback targets. • Configure and maintain "Vector Databases" and RAG pipelines, optimizing retrieval performance and index freshness. • Implement "Prompt Filtering" and content moderation gateways to prevent jailbreaks and enforce safety standards at the infrastructure level. • Develop "Blue/Green" or "Canary" deployment strategies for AI agents to safely test new model versions in production. • Manage the "API Gateway" for all AI services, ensuring authentication, rate limiting, and usage logging are enforced. • Optimize "GPU/CPU Orchestration" to control compute costs while maintaining performance SLAs for high-volume inference. • Build automated "Drift Detection" alerts that trigger retraining or human review when model performance degrades below a set threshold. • Perform any other job related duties as requested.