• Participate across the AI solution lifecycle from requirements and design through development, testing, and deployment • Build and maintain systems that exercise agentic workflows against realistic scenarios, edge cases, and failure conditions • Define contracts, validation, and explicit failure behavior at integration points • Create shared libraries, reference harnesses, and validation utilities that teams can reuse • Architect agentic systems so correctness, safety, and expected behavior are enforced by design • Participate in solution design to identify likely failure patterns in agentic systems • Define and implement cost-conscious evaluation strategies for AI-enabled systems • Work with cross-functional program teams to translate customer and mission needs into concrete evaluation criteria • Communicate estimates, progress, and technical tradeoffs clearly