gemini-embeddings

A real-time, AI-powered interview assistant that connects with platforms like Google Meet and Zoom to transcribe live interviews using Deepgram’s SST API (~100ms latency). It extracts questions, retrieves context from user-uploaded resumes or documents (via Gemini + Pinecone), and generates accurate responses using Claude.

Gemini Pro LLM and Pinecone Vector Database for fast and performant Retrieval Augmented Generation (RAG) with LlamaIndex

AI-powered customer support chatbot using n8n workflows with RAG (Retrieval-Augmented Generation), Groq LLM, Gemini embeddings, and vector store for dynamic FAQ retrieval

talk2pdf is an AI-powered application that enables seamless, multilingual voice and text interaction with your PDFs. It combines advanced retrieval-augmented generation (RAG), Gemini AI, and speech APIs to support natural, conversational, and voice-based queries in multiple languages, making document exploration simple and interactive.

chatgpt-project

A plug-and-play framework for a RAG (Retrieval-Augmented Generation) pipeline using Google's Gemini Embedding 2 model, the first fully multimodal embedding model to store and query embeddings across text, images, video, and audio.

A medical chatbot that delivers health information, answers queries, and assists in diagnosis using RAG-based with a powerful language model for improved accuracy and reliability.

backend for Rag pdf and image understanding projects

A search engine for research papers, built by students for students

VerbaVista is an LLM-powered Streamlit app that transforms any YouTube video into structured English notes and an interactive chatbot. It automatically fetches, translates, chunks, and embeds transcripts using Gemini + LangChain, enabling contextual Q&A through a RAG pipeline with Chroma.

An AI-driven learning tool using RAG to analyze web, YouTube, and text content. Features a 3D Three.js UI with glassmorphism, powered by Flask, FAISS, and Gemini API. Offers Q&A, summaries, mock tests, and visualizations like mind maps and flowcharts for an engaging educational experience.

Self-hosted memory for LLM agents - one SQLite file, zero services, MIT. Hybrid retrieval (FTS5 + sqlite-vec + RRF) with pain-weighted salience and a knowledge graph. Under matched embeddings, outperforms Mem0 on LoCoMo and LongMemEval. Your data, your provider, your disk.

GenAI | RAG-based system for semantic Q&A over YouTube transcripts using FAISS, Gemini embeddings, and LLM-driven context-aware generation

A CV Scorerer built with React and FastAPI. Uses Gemini API and Gemini Embeddings.

An intelligent medical document analysis system that uses AI to help users understand and query their medical PDFs through natural language conversations.

Graphiti service for manufacturing assets.

An AI-powered music classifier and semantic search tool built with Google Gemini Embeddings 2 and Electron.

🔍 100%-offline-capable multimodal semantic search for macOS. Find your files by meaning, not name — text, code, PDFs, images, audio. Ollama + Qwen3-VL local, Gemini/OpenAI optional. Power-user tool, macOS-only.

Multimodal semantic search for local photo libraries using Gemini Embedding 2 — search by text or image, no captions required.

Production-grade Natural Language to SQL (NL-to-SQL) assistant powered by FastAPI & React 19. Features a 5-stage self-healing pipeline, dual-tiered semantic cache (ChromaDB + Gemini), auto golden query generation, and chunked Doc-QA ingestion with exact page-level PDF deep-linking.