⚡️Framework for fast persistent storage of multiple document embeddings and metadata into Pinecone for source-traceable, production-level RAG.

A product knowledge base powered by Pinecone API and Anthropic AI Copilot

The goal of this application is to generate suggestions based on the given resume of the candidate, store the candidate profile in Pinecone database, and shortlist candidates accroding to the skills matched with match score.

SmartRAG-Assistant/GenAI-Assistant leverages advanced LLM models and Nvidia APIs for efficient query handling and document summarization. It integrates LlamaParse for structured data extraction, HuggingFace embeddings for vectorization, and PineconeDB for efficient retrieval, ensuring precise answers to user queries.

A chatbot designed to provide students with the best professors to match their needs through simple queries. The AI effectively uses a RAG implementation to provide accurate results.

A simple AI-based "Rate My Professor" using Next.js, OpenAI, and Pinecone for easy professor reviews and ratings.

A Question-Answering chatbot built using RAG (Retrieval-Augmented Generation) with conversation memory. This project uses LangChain, various LLM options, and vector stores to create an intelligent chatbot that can answer questions about Jessup Cellars winery.

A user-friendly RAG-powered fitness assistant — a conversational AI that understands your fitness goals, experience level, and equipment availability. It can help you select the perfect exercises, suggest alternative options, and keep you motivated to stay consistent with your routine, making fitness more accessible and personalised.

GenAI: Build and deploy end to end medical chatbot

Second Brain API: A backend service for managing, searching, and sharing personal content with secure authentication, robust validation, embedding-powered queries, and shareable links, built using Node.js, MongoDB, and Pinecone.

Movie Recommendation System: A content-based recommendation platform built with Python, Pinecone, and Streamlit. The system provides personalized movie suggestions based on genres and metadata, allowing users to explore tailored recommendations. With interactive genre filtering & clean interface, the app enhances movie discovery , hosted on render.

An End-to-End Medical Chatbot powered by generative AI, designed to provide accurate responses to medical queries. Built using Flask, Cohere’s Language Model, and Pinecone for Vector Storage.