Design and deliver production generative AI systems: RAG pipelines, agent workflows and governed AI APIs integrated with enterprise platforms.
Responsibilities
- Architect RAG pipelines covering ingestion, chunking, embeddings, retrieval and reranking
- Build agent workflows with typed tool registries and validation gates
- Design evaluation harnesses and guardrails ahead of production rollout
- Deploy AI services on cloud infrastructure with monitoring and cost controls
Major achievements
- Established a reusable AI service pattern adopted across multiple internal applications
- Introduced evaluation-before-rollout as a standard delivery gate
- Reduced integration effort for new AI use cases through a shared gateway layer
Technologies
PythonLangChainFAISSFastAPIAWSAzureDocker