Jayalakshmi Mattummal

I built a RAG PDF chatbot that lets users ask natural-language questions about uploaded documents. I implemented the pipeline from PDF ingestion and text chunking through semantic retrieval and LLM-generated answers, then deployed it on Streamlit Cloud.

For the chatbot, I used LangChain, Llama 3.1 through Groq, HuggingFace Embeddings, and FAISS. I also managed API keys with environment secrets.

I developed a Flask-based decision-support application that compares alternatives using data normalization and heuristic scoring. I converted a Python CLI prototype into a web application with transparent recommendations that don't rely on generative AI.

At Cognifyz Technologies, I cleaned, analyzed, and visualized data using Python, Pandas, and Tableau. I’m currently contributing to backend web development with Python and Django at Zecser Business, and I’m seeking to apply my skills as a Management Trainee or Associate Engineer in Generative AI.