I build production-ready, LLM-driven applications that solve real-world problems. With a strong foundation in Computer Science (9.1 CPI), I specialize in designing and deploying systems using Python, FastAPI, LangChain, and RAG architectures. My focus is on creating scalable solutions, from document intelligence to generative AI APIs.
- 🤖 Building robust RAG pipelines with tools like LangChain, FAISS, and Google Gemini.
- 🚀 Deploying applications using Docker, Render, and GitHub Actions.
- 💡 Passionate about optimizing model performance and minimizing API latency.
| Category | Technologies |
|---|---|
| Languages & Frameworks | Python, FastAPI, PyTorch, SQL, Streamlit |
| AI / ML & LLMs | LangChain, Hugging Face, FAISS, Gemini API, Sentence Transformers, RAG |
| Databases & Storage | PostgreSQL, MySQL, MongoDB |
| DevOps & Tools | Docker, Git, GitHub Actions, Render |
| Project | Description | Tech Stack |
|---|---|---|
| WordForge | LLM-powered blog generator with a FastAPI backend and PostgreSQL for content management. | FastAPI, PostgreSQL, Google LLM, Docker |
| SentinelRAG | Security log copilot using a RAG pipeline that improved query relevance by 30% with <780ms retrieval. | LangChain, FAISS, Sentence Transformers |
| DocuMind | Multi-file document QA bot processing 45+ pages with <10s response latency and 85% relevance. | LangChain, FAISS, Streamlit, Google LLM |
| ResumindAI | Resume analyzer providing instant scoring and ATS feedback with a ~85% success rate in text extraction. | Streamlit, Google LLM, PyPDF2 |
- Portfolio: pawansharma22.vercel.app
- LinkedIn: linkedin.com/in/pawansharma09
- Email: pawansharma91790@gmail.com