Yash Nayi

I am an AI/ML Engineer specializing in building production-grade LLM applications, agentic AI systems, and RAG pipelines that solve real-world problems at scale. With a Master’s in Data Science (GPA 3.7) and over 2.5 years building AI-powered backend systems, I focus on bridging research and production—turning cutting-edge models into reliable, business-impacting products.

I design multi-agent AI systems with tool use, structured outputs, and multi-step reasoning for enterprise workflows; build retrieval-augmented generation pipelines using Pinecone, FAISS, and other vector databases to reduce hallucinations and improve grounding; and deploy low-latency inference APIs via FastAPI, serverless architectures, and containerized microservices on AWS.

I deploy and fine-tune LLMs (LLaMA, Mistral, Claude, and others) using techniques such as LoRA/QLoRA, prompt engineering, and model optimization, and implement MLOps best practices including CI/CD pipelines, experiment tracking, model versioning, observability, and human-in-the-loop feedback to continuously improve system performance.

Before specializing in generative AI, I built scalable backend services with Django, Flask, and PostgreSQL—optimizing APIs, data pipelines, and database queries that served hundreds of concurrent users and materially improved system responsiveness.