I design RAG pipelines and engineer AI systems for efficient knowledge retrieval and generation, turning data into intelligent, context-aware solutions.
- RAG Systems — Chunking strategies, embeddings, and context-aware retrieval for efficient knowledge access
- LLMs — Designing and deploying large language model–based applications
- Latency Optimization — Improving inference speed and overall system efficiency for real-time AI applications
- MLOps & Deployment — Building, evaluating, and deploying scalable and reliable AI systems in production
LLMs & AI: OpenAI, LangChain, HuggingFace
Backend: Python, FastAPI, REST APIs
Databases: PostgreSQL, MongoDB, Vector DBs (FAISS, Chroma)
Tools: Git, Docker, Postman
| Project | Description | Stack | Impact |
|---|---|---|---|
| AI Research Assistant | AI-powered research assistant leveraging RAG pipelines and LLMs to help researchers analyze papers, retrieve knowledge, and accelerate academic writing. | RAG, LangChain, FAISS | Knowledge retrieval |
| Lung Adenocarcinoma Survival Prediction | A research-based project focused on predicting survival outcomes for lung cancer patients using machine learning models. | Machine Learning | Predictive insights |
- Context → Retrieve → Reason → Act → Evaluate
- Guardrails before execution
- Optimize retrieval quality continuously
- Track latency + cost metrics