Rushikesh Nimkar's portfolio, accessible at rushikeshnimkar.xyz, is a modern website built with Next.js 15 and TypeScript. It features AI-powered email generation, an interactive chat with an AI version of Rushikesh, dynamic animations, responsive design, dark mode, and showcases his projects and contributions.

A fast semantic search and retrieval engine

A space for learning GraphRAG with Microsoft Agent Framework and Neo4j, with a Blazor front end orchestrated by Aspire. Downloads, splits and ingests PDF data into Neo4j, and queries it using an agent workflow. Uses models from Azure Foundry as well as local Ollama models.

GenAI prototypes for Microsoft Agent Framework and Semantic Kernel: chat completions, tool calling, document ingestion, vector search, MCP, Azure Open AI, GitHub Models, Ollama and AWS Bedrock

FAIZ AI 🔍 – The search bot that finds what others miss. Searches HTTP, FTP, IPFS & Torrent with semantic vector embeddings + deep-web agents. Features intent/spell correction & file safety verification. ⚡ Python CLI tool. If you search deep web tools, semantic search, or multi-protocol crawlers

This is a node.js project leveraging the OpenAI API for vector embedding, seamlessly integrating with MongoDB to store embedded data and facilitating efficient query-based retrieval for enhanced knowledge management

This AI tool helps you to chat with your PDFs just by uploading it in the web interface.

RAG_llama3.3 is an advanced Retrieval-Augmented Generation (RAG) system using Llama 3.3 language model. This project integrates state-of-the-art natural language processing (NLP) techniques to enable accurate and context-aware question answering.

An AI-powered web scraper built with Crawl4AI that crawls websites, extracts and embeds text using HuggingFace, performs semantic search with FAISS, and answers queries using Langchain and GROQ LLM - all through a simple Gradio UI.

A JavaScript-based personalized AI system with dual-memory architecture, real-time conversation analysis, and secure chat isolation for intelligent, context-aware interactions.

LLM Learning step-by-step.

llama2 flask api for generative AI

A production-ready AI agent pipeline using LangChain for LLM workflows, LangGraph for agent orchestration, and LangSmith for observability supporting document processing, weather queries, and general Q&A with built-in monitoring and debugging.

LangChain Chatbot is a conversational AI system designed to assist users with legal queries and provide relevant information. It utilizes various natural language processing techniques, including OpenAI's GPT-3.5 model, Sentence Transformers, and Pinecone indexing, to understand user queries, refine them, and find the most relevant responses.

End-to-end retrieval-augmented question answering pipeline with embeddings, FAISS vector search, and FLAN-T5 generation.

RAG System QA for Techincal book

A multimodal RAG system that extracts text + images from PDFs, generates CLIP embeddings, stores them in FAISS, and answers queries using LangChain and a local LLM.

Built an AI-based multi-document chatbot using Retrieval-Augmented Generation (RAG) that enables conversational querying of PDFs with semantic search.

A lightweight image search engine that lets you upload a dataset and find similar images using either text queries or image queries, powered by CLIP embeddings, FAISS vector search, FastAPI, and a Streamlit UI.