AI/ML engineer (B.Tech in AI & ML, 2026) based in Noida, India. I build LLM applications that show their work: retrieval-grounded answers, page-level citations, and an honest "I don't know" when the source is silent.
| Project | What it does | Stack |
|---|---|---|
| Grounded RAG Engine | Answers questions strictly from uploaded PDFs, with file, page and snippet citations. Includes a retrieval gate, a strict refusal prompt and server-side refusal detection to reduce hallucination. | Python, FastAPI, LangChain, ChromaDB, Groq |
| Story-Agent | AI agent that generates interactive stories from user prompts. | Python, JavaScript, LLM agents |
| VectorShift-AI-Automation | GenAI orchestration and automation pipeline built for a VectorShift technical assessment. | Python, VectorShift |
- Languages: Python, JavaScript, Java, C++, SQL, HTML, CSS
- AI / ML: LLMs, RAG, NLP, embeddings, vector search, prompt engineering, PyTorch
- Frameworks & tools: FastAPI, LangChain, ChromaDB, Hugging Face, NumPy, Pandas, Node.js, React, Bootstrap, Git, GitHub
Ask me about: RAG, LLM guardrails, FastAPI, DSA.