crag

The bedrock layer for AI coding agents. One governance.md. Any project. Never stale. Universal skills + cross-agent compilation (Claude, Cursor, Codex, Gemini, Aider).

Implementation of Corrective RAG using LangChain and LangGraph.

It shows how to use langgraph-builder.

Training code for advanced RAG techniques - Adaptive-RAG, Corrective RAG, RQ-RAG, Self-RAG, Agentic RAG, and ReZero. Reproduces paper methodologies to fine-tune LLMs via SFT and GRPO for adaptive retrieval, corrective evaluation, query refinement, self-reflection, and agentic search behaviors.

Nous: A privacy-focused personal knowledge assistant using local LLMs to securely interact with your documents and enhance information retrieval.

Customer Driven Project (TDT4290) The project is a collaboration between NTNU and KartAI. Showcasing use of AI for streamlining building case process

RAG system using Hugging Face models, multiple vector stores (Chroma, Pinecone, FAISS), and CRAG, with sentence transformers and benchmarking tools for optimized retrieval and content generation.

🧮 Multi-agent AI math tutor built with LangGraph — CRAG retrieval, episodic & semantic long-term memory, Tavily MCP web search, Google OAuth, and Neo4j-style memory graph. Powered by LLaMA 3.3 70B on Groq.

Production-style RAG learning project — trace the full pipeline from ingestion to hybrid retrieval, RRF, reranking, and LLM answers. Java 21, Spring Boot, pgvector & Docker.

Investigating the efficacy of Retrieval-Augmented Generation (RAG) and Corrective Retrieval-Augmented Generation (CRAG) in harnessing external knowledge to improve AI model performance and output quality.

Autonomous multi-agent due-diligence desk — LangGraph orchestration, agentic hybrid RAG, MCP tools, and a self-improving LLM-Ops loop.

DecisionLens: A Corrective RAG Engine for IFAB/FIFA Rule Compliance and VAR Transparency

Production-grade RAG Document Intelligence Platform — LangGraph Adaptive RAG + CRAG, LangSmith observability, RAGAS evaluation,Qdrant hybrid search (BGE dense + BM42 sparse + RRF), jinaai/jina-reranker-v1-tiny-en  re-ranking, FastAPI + Docker

Imagine a website where users can skip complex navigation and get instant answers with just a question. This project explores how a Corrective-Retrieval-Augmented Generation (CRAG) chatbot reduces server load and network congestion by streamlining interactions-enhancing both efficiency and user experience in a way traditional navigation can't.

RAGWarrant: evidence-preserving governance and promotion control for RAG policy changes.

A repos containing LangGraph learning with end to end project

Corrective Retrieval Augmented Generation (CRAG) with OpenEvals on LangGraph

A Small Collection of Python Games.