R

RAG

A complete RAG demo application in Python with a multi-stage pipeline to minimize hallucinations.

These projects form an open-source suite of AI infrastructure tools built for modularity, security, and self-hosted deployment, allowing users to maintain full ownership and control of their systems and data. They deliver interoperable AI functions including automation, data retrieval, encryption, and identity management that can be applied across many different industries.

Lightweight vector embeddings store for RAG applications

Curated retrieval-augmented generation (RAG) frameworks, tools, and reference implementations.

Curated vector databases, vector search engines, and embedding storage systems for AI.

ApolloCore is a self-contained, embeddable knowledge base that runs 100% locally with no external API dependencies. It combines vector search, keyword search, knowledge graphs, entity extraction, and LLM-powered query enhancement into a single high-performance C++23 library.

🧠 Open-source persistent memory engine for AI agents and LLMs with semantic search, automatic deduplication, and intelligent context retrieval.

A persistent-memory AI agent with cryptographically signed, hash-chained, tamper-evident memory. Pre-commit quality scoring quarantines prompt-injection; protected zones guard core records. Scans its own history for recurring gaps and proposes improvements for human review. Provider-agnostic (Claude, GPT, Gemini, DeepSeek, OpenRouter, Ollama).

RAGBase is a schema-enforced infrastructure layer for building modular Retrieval-Augmented Generation (RAG) systems. It provides a strict, contract-driven architecture where every component—ingestion, embedding, retrieval, and generation—is defined by explicit input/output schemas and validated through a built-in spec hygiene layer. This ensures predictable behavior, prevents system drift, and enables safe module interchangeability in production-grade AI pipelines.

Production-ready RAG starter: hybrid search, chunking strategies, observability (Prometheus/Grafana), MLflow tracking, drift detection, GDPR deletion, and evaluation. The parts the tutorials skip.

RAG-Powered SOC Assistant - By Ayi NEDJIMI

https://ayinedjimi-consultants.fr

GPU-accelerated embedding server for RAG systems - CUDA, FastAPI, sentence-transformers | Serveur d'embeddings GPU ultra-rapide

https://ayinedjimi-consultants.fr

Yanapa - Asistente AI offline para comunidades originarias de Sudamerica y Centroamerica. Salud, derechos, cultura y medicina tradicional en lenguas indigenas.

Hybrid credit intelligence platform using XGBoost, RAG, Terraform and GitLab CI/CD.

⚡ Build structured YouTube datasets at scale — effortlessly fetch transcripts and rich metadata for NLP, ML, and AI workflows.

An AI-powered tactical inteligence system using RAG, FastAPI and vector search.

Project Description: Puran Story to Animation Generator

This project is an AI-powered storytelling and animation system that transforms user prompts into narrated mythological stories and animated videos based on Hindu Puranic texts.

The system uses a local Large Language Model (LLM) to generate context-aware stories with references from curated Puran datasets. It then converts these stories into visual scenes, generates images, adds voice narration, and compiles everything into a video animation.

Key Features:

🧠 Domain-specific LLM trained on Hindu Purans 🔍 Semantic search (embeddings + vector DB) for accurate context retrieval ✍️ Story generation with citations 🎬 Automatic scene breakdown from story 🎨 AI-based image generation for each scene 🔊 Text-to-speech narration 🎥 Video creation using generated visuals + audio 💻 Interactive UI using Streamlit

🧩 Tech Stack:

LLM: Local models via Ollama Embeddings: Sentence Transformers Vector DB: ChromaDB Image Generation: Diffusers (or Stable Diffusion) Audio: pyttsx3 / TTS Video: FFmpeg UI: Streamlit

🎯 Workflow:

User Prompt → Retrieve Context → Generate Story → Split into Scenes → Generate Images → Generate Voice → Combine → Video Output

💡 Purpose:

The project aims to: Make ancient Puranic knowledge more engaging Combine AI + storytelling + visualization Provide an interactive learning and entertainment tool

Flexible GraphRAG: Python, LlamaIndex (LangChain also coming) Docker Compose: 8 Property Graph dbs, 3 RDF graph dbs, 10 Vector dbs, OpenSearch, Elasticsearch, Alfresco. 13 data sources (9 auto-sync), KG auto-building, RDF ontologies, schemas, LLMs, document processing, Docling, LlamaParse, GraphRAG, RAG only, Hybrid search, AI chat. React, Vue, Angular TypeScript frontends, FastAPI backend, REST API, MCP Server

Multi-source RAG pipeline with hybrid vector + keyword retrieval, LLM-powered concept knowledge graph, adaptive search weighting, and evaluation framework.

An intelligent chatbot that answers user queries strictly based on an uploaded document (PDF/DOCX) using a Retrieval-Augmented Generation (RAG) pipeline.