High-accuracy PDF-to-Markdown OCR API using LLMs with vision capabilities. Features parallel processing, batching, and auto-retry logic for scalable extraction.

A Python CLI to test, benchmark, and find the best RAG chunking strategy for your Markdown documents.

RAG boilerplate with semantic/propositional chunking, hybrid search (BM25 + dense), LLM reranking, query enhancement agents, CrewAI orchestration, Qdrant vector search, Redis/Mongo sessioning, Celery ingestion pipeline, Gradio UI, and an evaluation suite (Hit-Rate, MRR, hybrid configs).

A python module library that simplifies RAG through abstraction

The implementation of Test Time Diffusion paper by Google with some tweaks to run on 24gb gpu

The Audited Context Generation (ACG) Protocol prevents AI hallucinations with a dual-layer system. The UGVP layer links every fact to a precise source for verification. The RSVP layer audits the AI's logical reasoning when combining facts. This creates a fully transparent, machine-auditable trail for both source and logical integrity.

An intelligent customer support system powered by LangGraph and LangChain that uses Retrieval-Augmented Generation (RAG) to provide accurate, context-aware responses to customer queries. Built with FastAPI, FAISS, and multi-stage validation for production-ready deployment.

Production-ready Chainlit RAG application with Pinecone pipeline offering all Groq and OpenAI Models, to chat with your documents.

Token-Oriented Object Notation (TOON) is an LLM-optimized data serialization format implemented in Python.

When retrieval outperforms generation: Dense evidence retrieval for scalable fake news detection - LDK 2025

AI travel planner with 7 specialized agents, RAG, and tool-calling. Built with CrewAI & LangChain. Generates personalized itineraries with flights, hotels, activities, and cultural tips. Production-ready Python codebase.

Advanced RAG Pipelines and Evaluation

🛡️ Web3 Guardian is a comprehensive security suite for Web3 that combines browser extension and backend services to provide real-time transaction analysis, smart contract auditing, and risk assessment for decentralized applications (dApps).

Demo LLM (RAG pipeline) web app running locally using docker-compose. LLM and embedding models are consumed as services from OpenAI.

This repo is for advanced RAG systems, each branch will represent a project based on RAG.

Advanced RAG Pipelines optimized with DSPy

Teaching Tool: Local RAG and LLM Chat over Logseq with LlamaIndex + Ollama + Chroma

Project Agora: An expert system for the Google ADK, powered by a hierarchical multi-agent framework to automate code generation, architecture, and Q&A.

AI-driven prompt generation and evaluation system, designed to optimize the use of Language Models (LLMs) in various industries. The project consists of both frontend and backend components, facilitating prompt generation, automatic evaluation data generation, and prompt testing.