A collection of projects that reflect my dedication to both craft and creativity.
This project is an AI-powered PDF chatbot that allows users to upload PDFs and ask questions based on the document's content. It uses LlamaParse to extract text, splits the content into chunks, and generates vector embeddings using OpenAI’s API. These embeddings are stored in a Weaviate vector database for efficient retrieval. A Retrieval-Augmented Generation (RAG) pipeline ensures the chatbot provides accurate, context-aware answers. The tech stack includes React, Node.js, Tailwind CSS, and Shadcn UI.
A full-stack recruitment platform that automates candidate interview analysis using AI/ML. The system evaluates facial expressions, speech emotions, and spoken responses in real-time to generate performance insights. It helps recruiters make unbiased, data-driven hiring decisions with detailed AI-generated reports.