Tag: RAG

Memoria brings Git-level version control to AI agent memory with snapshots, branches, and rollback. Built on MatrixOne's Copy-on-Write engine, it eliminates hallucinations and enables safe experimentation.

Personal-Graph is an open-source Python library that gives AI systems real memory through knowledge graphs. Built on libsql with natural language interfaces, per-user isolation, and ML-ready exports. Stop losing context—start building intelligent memory.

Discover EgoAlpha's daily-updated open-source hub for prompt engineering mastery. 1000+ curated papers, annotated LangChain tutorials, and production-ready techniques for ChatGPT, GPT-4, and beyond. Stop wasting tokens—start engineering prompts that work.

ReadAny is an open-source AI-powered e-book reader that combines semantic search, RAG-based AI chat, local vector storage, and cross-device sync. Built with Tauri 2 and React 19, it transforms how developers read, annotate, and retain knowledge from technical books.

HelixDB is the open-source graph-vector database built from scratch in Rust that unifies your AI backend. Replace your fragmented stack of vector DB, graph DB, and application layers with one system.

Discover awesome-context-ai: the definitive curated list of 50+ tools for building AI with rich context from screen recordings, audio, and personal data. Stop building forgetful AI—start building systems that truly understand your users.

CocoIndex is an open-source incremental data transformation framework with a Rust core that keeps AI agent context fresh. Only process the delta — stop wasting compute on full re-embeddings.

Microsoft's free 21-lesson generative AI course on GitHub teaches developers to build production-ready AI applications with Python and TypeScript. Covers LLMs, RAG, agents, fine-tuning, and more.

OpenDataLoader-PDF is the #1 benchmark-ranked open-source PDF parser for AI-ready data extraction and accessibility automation. Achieves 0.907 overall accuracy, generates Tagged PDFs end-to-end, and runs entirely local with zero cloud dependency.

SearchArray embeds BM25 and TF-IDF search indexing directly into Pandas DataFrames and NumPy arrays. Stop over-engineering with Elasticsearch for small-to-medium datasets—search with .score() like it's .mean(). Complete guide with real code examples.

Discover Endee, the open-source vector database handling 1 billion vectors on a single node. Built in C++ with AVX2/AVX512/NEON optimizations for AI search, RAG pipelines, and agentic AI memory systems.

Discover how Zep's sub-200ms context engineering platform uses temporal knowledge graphs to transform AI agent reliability. Complete setup guide, real code examples, and architectural deep dive included.

Discover how Polymarket Agents enables autonomous AI-powered trading on prediction markets. Learn installation, architecture, real code examples, and advanced strategies for building intelligent trading agents with LLM integration and RAG support.

Discover Crawl4AI, the viral 50k+ star open-source web crawler transforming pages into LLM-ready Markdown. Free, production-hardened, and built for AI pipelines. Complete guide with real code examples.

Discover how Chainlit lets Python developers build production-ready conversational AI applications in minutes instead of weeks. No frontend code required — just pure Python decorators, instant hot-reload, and automatic step visualization.

GBrain is Garry Tan's open-source AI agent memory system—17K+ pages, self-wiring knowledge graph, hybrid search, and 34 production skills. Install in 30 minutes, get persistent agent intelligence that compounds overnight.

Discover how RAGatouille makes state-of-the-art ColBERT retrieval accessible for any RAG pipeline. Stop struggling with weak dense embeddings and unlock research-grade retrieval with simple pip install.

Discover PaddleOCR, the 70k+ star open-source toolkit transforming PDFs and images into structured LLM-ready data. With 100+ language support, specialized vision-language models, and deep RAG ecosystem integration, it's the production-ready alternative to fragmented OCR pipelines.