AI/ML practitioner who builds end-to-end intelligent systems - from raw data to production-ready AI.
I've designed RAG pipelines, integrated LLM APIs and built machine learning models that solve real business problems, not just notebook demos.
2+ years across data analytics and AI/ML engineering:
Data Analyst, HailCabs - a fast-paced mobility startup where data decisions moved fast and ownership was real. Built SQL pipelines and Power BI dashboards, and owned operational reporting end-to-end. Cut manual reporting effort by 60% and improved decision speed by 30% and AI Engineer Intern, Infosys (1 year, on-site) - built LLM-powered applications, RAG pipelines and intelligent data extraction systems using Python, LangChain and OpenAI APIs, working in Agile sprints alongside senior engineers.
Selected build: a RAG chatbot (LangChain, FAISS, OpenAI API, Streamlit) that answers queries over medical research and clinical documents using semantic search.
Certified: Infosys GenAI Professional (Advanced), Infosys ML Professional, Infosys Python Associate, Meta Data Analyst Professional Certificate.
| Area | What I Do |
|---|---|
| π€ Generative AI & LLMs | LLM API integration, prompt engineering, RAG pipelines, output optimization |
| 𧬠Machine Learning | End-to-end ML β preprocessing, feature engineering, model training & deployment |
| π‘ AI Backends | FastAPI-based AI microservices, REST APIs, data pipeline architecture |
| π Data Analytics | Business intelligence, SQL analysis, Power BI dashboards, data modeling |
| π€ AI Agents | Autonomous agent workflows with LangChain, tool use, multi-step reasoning |
- π€ AI Agents & Agentic Workflows - LangChain Agents, CrewAI, tool-calling
- ποΈ Vector Databases - Pinecone, ChromaDB, FAISS for semantic search & RAG
- βοΈ MLOps - Model serving, CI/CD for AI, AWS SageMaker
- π§ Advanced LLM Architectures - Fine-tuning, RLHF, transformer internals
π― Actively looking for AI Engineer Β· ML Engineer Β· Generative AI Β· Data Scientist roles
π‘ Building AI solutions Β |Β π Data-driven mindset Β |Β π Always learning!

