vector-search

Minimal RAG (Retrieval-Augmented Generation) example using Sentence Transformers and Qdrant for semantic search over text files

RecallOps is an agentic incident-memory system that stores verified outcomes in CockroachDB, recalls resolved precedents with vector search, audits evidence through Managed MCP, and grounds recommendations with Amazon Bedrock.

AI-powered Retrieval-Augmented Generation (RAG) system that transforms websites into a searchable knowledge base using Playwright, OCR, OpenAI Embeddings, MongoDB Atlas Vector Search, hybrid retrieval, and LLM reranking.

Embedding Search sample with Voyager

Self-hostable, web-accessible semantic video search. Embed and search your video library with natural language.

My Experiments with LLM augmentation using RAG, Vector embedding's, Vector DB store, Vector Models, Agentic Frameworks, and bit of typescript todo it all.

Clinical decision support with BigQuery AI (semantic search + AI-generated care cards)

Agentic RAG over the LangChain/LangGraph/LangSmith docs: a LangGraph state machine that plans queries, runs hybrid retrieval (FAISS + BM25 + Cohere rerank), routes to specialized agents, and verifies answers against sources.

Querious – Chat with your documents using retrieval-augmented generation (RAG). Upload PDFs, ask questions, get grounded answers.

Vector Database Setup, unstructured text document ingestion and Search - Small sample

Retrieval-Augmented Generation (RAG) implementation for a Q&A bot for documents

This application demos Couchbase vector search of restaurants in California stored in Capella. Enabling users to search for more diverse exploration e.g. sushi restaurants with a good vibe and unique tasting menus. Beats OpenTable filters.

A Retrieval-Augmented Generation (RAG) chatbot built with LangChain and OpenAI, using FAISS for vector search over PDF documents. Includes a Streamlit UI and LangSmith tracing for query monitoring.

Efficient Serving of Large-scale Vector Search with Sharded Indexes

This app allows users to search for products by either entering text or uploading an image, and retrieves relevant products from a database

Hybrid RAG service that cuts p50 query latency 91.9% under concurrent load (IVF ANN + LRU cache vs exact search, 2 vCPU) and lifts precision@5 from 0.839 to 0.863 via FAISS + BM25 reciprocal rank fusion. LangGraph supervisor with guardrail, retriever, and synthesizer agents; FastAPI, 95% test coverage, reproducible committed benchmarks.

Production-ready AI Agent Memory library for Python, FastAPI, LangChain, LangGraph, MCP, and LLM applications. Supports Redis, PostgreSQL, SQLite, vector databases, semantic memory, session memory, RAG, caching, and enterprise-scale AI workflows.

Educational & Geospatial Retrieval-Augmented Generation System. A hybrid RAG and agentic system built in Python for answering natural language queries about Brazilian educational and geospatial data (based on Cultura Educa and INEP datasets).

A retrieval-augmented generation agent that performs hybrid search over documents and generates answers using large language models.

🌊 TypeScript RAG library with knowledge graph support — batch indexing, semantic search, entity extraction, and graph traversal. Lambda-friendly, Git-friendly, zero servers.