Madhavi latha J

I’m a detail-oriented and analytical Machine Learning Engineer with 2+ years of hands-on experience in building scalable AI workflows and deploying LLM-powered applications. I focus on designing RAG pipelines, prompt engineering, and backend services that turn model outputs into reliable user-facing experiences.

In my current role at Cognizant Technology Solutions, I design and maintain scalable data pipelines and retrieval-augmented generation (RAG) workflows for diverse datasets. I develop and deploy LLM-powered applications using LangChain and FastAPI, integrating vector databases, and I engineer prompts plus evaluation frameworks to optimize latency, cost, and quality.

I also bring production discipline: I collaborate with cross-functional teams to translate requirements into AI features with observability and fallback mechanisms, and I deploy models using Docker and Kubernetes on AWS. I’m passionate about shipping robust, scalable, production-grade AI—whether through advanced agentic patterns like tool calling, planning, and memory management, or through rapid prototyping to deployment.