• Design and implement multi-step, agentic AI workflows (planner/executor, tool-using agents, RAG + memory patterns) • Engineer robust contextual pipelines including: • Retrieval-augmented generation (RAG) • Structured memory (short-term, long-term, episodic) • Tool invocation and function calling • Control hallucination, drift, and overreach through grounding, validation, policy constraints, and ensemble methods. • Optimize context windows, token usage, and latency under federal cost, platform quotas, and performance constraints. • Build production-grade services using Python, Java, C++, Rust, GO, C#, TypeScript • Implement APIs and microservices that integrate AI agents with: • Data lake house • SharePoint • Microsoft Graph API • Line-of-business systems • Apply secure coding practices aligned with NIST 800-53 / 800-171 • Design and deploy solutions in GCCH environments • Work within constraints such as: • Limited model availability • Restricted outbound network access • Controlled identity and authentication flows • Use Azure services including: • Azure AI (OpenAI / Foundry where applicable) • Azure Kubernetes Service (AKS) • Azure Functions / App Services • Azure Data Lake Storage • Azure Key Vault • Implement identity, RBAC, and managed identities aligned with federal requirements • Design AI systems with auditability, traceability, and explainability • Support Authority to Operate (ATO) processes • Implement logging, monitoring, and human-in-the-loop controls • Ensure data residency, data minimization, and model usage compliance