Udemy
    •  
    •  
    •  
    •  
    •  
    •  
    •  
    •  
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
Hands-On RAG with LangChain: Build Real-World Projects
Role Play
Rating: 4.7 out of 5(223 ratings)
3,654 students

Hands-On RAG with LangChain: Build Real-World Projects

Master Reranking, Vector Indexes, Caching, Ragas Evals & Metadata Filters for Production-Ready RAG
Created byBharath Thippireddy
Last updated 9/2026
English
English [Auto],Spanish [Auto],

What you'll learn

  • Set up PostgreSQL with the pgvector extension to enable efficient vector search
  • Build an end to end RAG pipeline connecting Large Language Models LLMs with PostgreSQL
  • Implement a RAG pipeline step-by-step: retrieval, context injection, and grounded LLM responses.
  • Explore vector indexes (HNSW, IVFFlat) and learn how they improve retrieval speed and accuracy.
  • Apply semantic caching to reduce cost and latency while improving response times.
  • Evaluate RAG systems using RAGAS metrics like faithfulness, context recall, and precision.
  • Enhance retrieval quality with re-ranking and metadata filtering.
  • Deploy APIs for ingestion and RAG Q&A to make your projects production-ready and testable.

Course content

9 sections • 54 lectures • 2h 44m total length

  • Introduction1:37

    Build a production-grade rag pipeline with Postgres pgvector for company knowledge base and ecommerce semantic search, including vectorization, chunking, ingestion, vector indexes, Redis caching, Evals, and reranking.

  • Private Course Feedback Link0:13
  • Software Setup3:29

    Install and configure tools and API keys for a hands-on rag project: Visual Studio Code, Docker Desktop, PGAdmin, OpenAI, Langsmith, and Cohere, enabling embeddings and LLM models.

  • Download Slides0:17

    Download a zip of the lecture resources, unzip to access the slides used in this course, including a Keynote file for Mac and a PowerPoint file for Windows.

  • Download Projects0:53

    Download the zip to access completed and skeleton projects, including company knowledge assistant. Drag the assistant into Visual Studio Code and follow shop wise ecommerce skeleton for the assignment.

  • Why PostgreSQL for Vector Search?2:17

    PostgreSQL with pgVector offers a unified vector store for rag applications, combining vectors, sql filters, and metadata in a reliable, open source production-ready database.

Requirements

  • Basic Python knowledge (functions, imports, virtual environments).
  • Familiarity with APIs and JSON is helpful but not mandatory.
  • An OpenAI API key
  • Docker basics are a plus, but we’ll cover the commands you need.
  • Curiosity to build real-world AI apps using vector databases and LLMs.
  • Visual Studio Code,Docker,PGAdmin

Description

Retrieval-Augmented Generation (RAG) is one of the most powerful ways to make Large Language Models (LLMs) smarter, more reliable, and production-ready. Instead of depending only on what the model “knows,” RAG allows us to fetch relevant knowledge from external sources and provide precise, up-to-date answers.

In this hands-on course, you’ll go beyond the basics and actually build RAG pipelines step by step using LangChain, the leading framework for LLM applications. Whether you are a developer, data scientist, or AI enthusiast, this course will give you the practical skills to design, implement, and optimize real-world RAG projects.


What You’ll Learn

  • Real-World Project: Build two end-to-end RAG Projects on Company Data and E-Commerce Semantic Search.

  • Caching Strategies: Use embedding and response caching to reduce cost, latency, and improve efficiency.

  • Indexing: Explore Flat, IVF Flat, HNSW, and disk-based indexes; learn which one to use for your dataset.

  • Reranking: Improve answer precision using similarity scores, cross-encoders, and LLM-based reranking.

  • Evaluations (Evals & Ragas): Measure faithfulness, relevance, and retrieval quality with Ragas metrics.

  • Metadata: Use metadata filters to make retrieval precise, context-aware, and production-ready.


Why Take This Course?

  • It’s hands-on — you won’t just learn theory; you’ll build working RAG pipelines.

  • You’ll learn best practices for scaling from demo to production.

  • Content is designed for real-world applications in enterprise, startups, and research.

  • You’ll walk away with code, skills, and confidence to build your own RAG-powered apps.

Who This Course Is For

  • Developers and data scientists interested in LangChain and LLM applications.

  • AI/ML engineers who want to deploy production-ready RAG systems.

  • Professionals curious about vector databases, embeddings, and retrieval systems.

  • Anyone who wants to go beyond ChatGPT and build AI that leverages their own data.

By the end of this course, you’ll have the knowledge and hands-on experience to design and implement efficient RAG pipelines with LangChain — and the skills to apply them to your own projects or business use cases.

Who this course is for:

  • Students who have completed my LangChain course and now want to dive deeper into advanced RAG use cases.
  • Python developers who want to integrate LLMs and vector search into real-world apps.
  • Data engineers & ML practitioners curious about Retrieval-Augmented Generation.
  • Backend developers exploring LangChain, RAG pipelines, and vector databases.
  • Anyone with basics of RAG and wants hands-on projects and not just theory.

Report abuse