{"repo":"T-Sunm/rag-ops","free":true,"listed":false,"github":"https://github.com/T-Sunm/rag-ops","clone":"git clone https://github.com/T-Sunm/rag-ops.git","description":"This project applies the core knowledge from the LLMOps module, including the design and implementation of the API Layer, Inference Layer, Observability Layer, Cache Layer, Guardrails Layer, Routing Layer, and the Data Ingestion Pipeline.","language":"Python","stars":74,"topics":["airflow","chatbot","chromadb","devops","langfuse","litellm","llm","llmops","minio","mlops"],"license":null,"category":"ai-agents","readme_excerpt":"End-to-End RAG Chatbot System with Langchain: From Ingestion to Guardrails Overview This project implements a complete Retrieval-Augmented Generation (RAG) chatbot system using Langchain and modern LLMOps best practices. It covers the full lifecycle of a RAG-based application — from ingesting documents and managing embeddings, to optimizing inference and ensuring observability, safety, and scalability. The system is designed in a modular and production-ready architecture, consisting of key layers such as embedding ingestion, inference, caching, observability, routing, and gateway. It supports both streaming and blocking APIs, integrates with tools like Langfuse, Airflow, vLLM, and FastAPI, and follows best practices in error handling, fallback strategies, and guardrail implementation. This project is ideal for developers, MLOps engineers, or researchers looking to understand, build, or deploy scalable and secure RAG systems in production. Features - Step-by-step RAG pipeline implementation using Langchain: ingestion, embedding, retrieval, and generation. - Modular architecture, with dedicated layers for: - API: Blocking + streaming (SSE/WebSocket), error handling, retries, and timeouts. - Embedding & Ingestion: Document loading, chunking, vector embedding, and storing using vector DBs (e.g., Milvus, Chroma). - Inference: Optimized high-throughput inference using vLLM and NVIDIA Dynamo. - Observability: Integrate Langfuse for prompt tracing, evaluations, and cost tracking. - C","default_branch":null,"files":null,"tree":[],"storefront":"/r/T-Sunm","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/T-Sunm/rag-ops/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}