AI Engineer · Building production-grade LLM systems
RAG pipelines · Agentic AI · LangGraph · AWS · FastAPI

Answers medical questions grounded in verified clinical documents using retrieval-augmented generation

• RAG pipeline — PDF ingestion → chunking → Pinecone vector search → Gemini LLM generation
• Deployed on AWS EC2 with full CI/CD via GitHub Actions → Amazon ECR → self-hosted runner
• Stack: LangChain · Pinecone · Google Gemini · FastAPI · Docker · AWS EC2/ECR

Generates high-quality blog posts using a multi-node agentic graph with self-evaluation and revision loops

• LangGraph agentic graph — title → outline → content → quality check → auto-revision loop
• Conditional routing — multilingual translation (Hindi, Marathi, French) via graph edges
• Tone control — professional / casual / academic / humorous writing styles
• Streaming API — real-time node-by-node progress via FastAPI StreamingResponse
• Stack: LangGraph · Groq LLaMA 3.1 · FastAPI · Streamlit · Render · LangSmith

AI / LLM

Backend & APIs

Cloud & DevOps

I focus on the gap between AI research and production systems — taking LLM capabilities and deploying them as reliable, observable, maintainable services:

• RAG systems that retrieve grounded context instead of hallucinating
• Agentic pipelines using LangGraph that loop, evaluate, and self-correct
• REST APIs with proper schemas, error handling, and streaming
• CI/CD pipelines that build, push to ECR, and deploy to EC2 automatically

I'm actively looking for AI Engineer / ML Engineer roles where I can work on production LLM systems.

• 💼 Currently at General Mills India
• 📍 Based in Mumbai, India
• 🔗 LinkedIn

"Ship it, then improve it."