• Design, implement, and evolve RAG pipelines combining structured data, embeddings, and LLMs. • Develop and maintain prompt strategies used across multi-step agent workflows. • Integrate LLMs into production systems with attention to reliability, cost, and latency. • Orchestrate AI-driven workflows across internal services and third-party APIs. • Define and improve practices for prompt versioning, evaluation, and iteration. • Work closely with platform engineers to ensure AI systems are maintainable and observable. • Help the team reason about when deeper customization or model-level work may be warranted in the future.