Engineered document parsing, extraction, and retrieval pipelines across enterprise AI applications, improving downstream answer relevance by 60% through optimized content processing and retrieval strategies. Built scalable backend services using Python, FastAPI, and asyncio, supporting asynchronous workflows, API integrations, and production AI/ML systems.
Managed and optimized data platforms spanning PostgreSQL, MongoDB, and vector databases (FAISS, Qdrant, pgvector), enabling efficient storage, retrieval, and processing of structured and unstructured data. Owned cloud infrastructure and deployment operations on AWS, utilizing EC2, S3, Route 53, Bedrock, Docker, and Docker Compose to support application hosting, model integration, and GPU-enabled inference environments.
Drove performance optimization initiatives, including integration of Go-based streaming services that increased concurrent stream processing capacity by 30% while improving resource utilization and system stability. Partnered with clients, product stakeholders, and cross-functional teams to gather requirements, analyze business challenges, and translate needs into scalable technology solutions delivered through Agile development practices.
