인프런 영문 브랜드 로고
인프런 영문 브랜드 로고
AI

/

Generative AI

RAG Master: From Basics to Advanced Techniques (feat. LangChain)

This course covers the basic concepts and implementation of RAG (Retrieval-Augmented Generation) systems using LangChain. Participants will learn to understand the core principles of RAG and how to build and evaluate the system in practice using LangChain.

(5.0) 7 reviews

194 students

RAG
LangChain
LLM
Chatbot
Python
Thumbnail

This course is prepared for Basic Learners.

What you will learn!

  • Building a RAG system using LangChain

  • Learn effective search and creation techniques

  • Performance Evaluation Method of RAG System

  • LangChain's LCEL Grammar and Runnable Usage

From RAG implementation to performance evaluation
Practical AI Development Completed in 9 Hours

Although the use of RAG systems has increased due to the advancement of AI technology, the process of learning and implementing them is still difficult. I also experienced many difficulties when I first studied RAG, and based on that experience, I planned this lecture.

The course focuses on building a working system based on step-by-step exercises, and covers not only basic implementations but also advanced techniques to improve search quality and performance evaluations, providing practical knowledge that can be applied directly to real projects.

Five months after launch
1590+ people took the course
Created the LLM series of lectures
RAG lecture by knowledge sharer

Python & Langchain
Free basic lectures provided
Python Basics Course
Langchain Basics Course

For building a RAG system
Rich learning materials
31 pages of summary data and
6 source code files

Lecture Points 💫

Building the foundation for RAG implementation

There is a limit to just writing code. You need to understand the principles of when and why you use this component . The lecture covers the basic concepts of RAG, major components, LCEL grammar, etc. , and builds the foundation for implementing RAG . We also provide a free Python & Langchain basic lecture for beginners.

Improve RAG implementation capabilities with the latest modules and techniques

The RAG process consists of [Document Load → Text Segmentation → Embedding → Vector Storage → Search → Prompt → LLM → Final Results]. The lecture introduces various cutting-edge modules and techniques applicable to each process . In particular, you can encounter advanced search techniques such as hybrid search, re-rank, and context compression to improve search performance.

10 Performance Evaluations for RAG Improvement

In order to improve the RAG system, the "evaluation-improvement" task is essential. In this lecture, we will introduce five methods of information retrieval to evaluate the search performance of RAG. We will also cover five methods of evaluating RAG's answers, including an evaluation method based on quantitative indicators and an evaluation method using LLM.

Easy and clear explanation power proven by numerous reviews

Many students have proven this

The course reviews are for courses opened by knowledge sharers as of September 24.

Learn about these things

Understanding the basic concepts of RAG and LangChain

Understand the operating principles of the RAG system, learn the structure of LangChain, and LCEL grammar. Through this, you will prepare a practical environment and acquire basic knowledge that can be utilized in various AI projects.

Hands-on : Installing LangChain, setting up the environment, and configuring a basic RAG pipeline

Practice data processing and text segmentation techniques

You can handle various data formats (PDF, JSON, Web, etc.) and learn effective text segmentation techniques to efficiently manage large-scale data.

Hands-on training : Hands-on training with various document loaders such as PyPDFLoader and CSVLoader, application of text segmentation strategies (recursive segmentation, utilization of regular expressions, semantic segmentation)

Utilizing embedding models and vector storage

By leveraging the embedding model, text data can be converted into vectors and stored in a vector storage to maximize RAG search performance.

Hands-on : Chroma, FAISS vector repository creation and search, OpenAI and Huggingface, Ollama embedding model utilization

RAG Search Performance Evaluation and Optimization

Evaluate RAG search performance using various information retrieval evaluation indices and acquire optimization techniques that can be applied to actual projects.

Practical training content : Search performance testing and evaluation (quantitative evaluation such as Hit Rate, MRR, etc.), optimization methods (query expansion, Re-rank, context compression) training

Generating and evaluating answers using LLM

You can generate answers in the RAG system by utilizing various LLMs, and evaluate the quality of the answers through the LangChain evaluation tool.

Practice content : LLM linkage practice, response evaluation using LangChain evaluation tool

Implementing a RAG-based chatbot using Gradio

Using Gradio, you can build a RAG-based chatbot interface that interacts with users, and design a real-time search and answer generation system.

What you will learn : Implementing a RAG chatbot using Gradio, stream-style output, and adding chat history

Things to note before taking the class

Practice environment

  • Operating System and Version (OS): Lecture will be based on MacOS (Linux and Windows users can also practice)

  • Using a virtual environment: The lecture will proceed based on Poetry (conda, venv users can also practice)

  • Tools used: VS Code, OpenAI API, etc. LLM authentication key required (separate cost may apply)

  • PC Specs: Not applicable

  • LangChain version: v0.2.16 applied

Learning Materials

  • Providing materials required for practice (lecture notes, practice code)

  • (For beginners) Reference material provided on Wikidocs: https://wikidocs.net/book/14473

Player Knowledge and Notes

  • People with basic knowledge of Python (those who can do basic programming)

  • LangChain Basics for Beginners (Free Course): https://inf.run/Fzfhs


  • If you have any questions or comments, please feel free to ask.

Linked lecture guide

  • RAG system implemented with AI agent (w. LangGraph)

  • Intelligent AI agent for augmented search generation (RAG) implemented with LangGraph


    • Design and implementation of AI agent structure using LangGraph

    • Applying AI Agents to Retrieval-Augmented Generation (RAG)

    • Expanding the capabilities of AI agents by implementing Tool Calling functionality

    • Mastering the latest agent RAG architectures including Adaptive RAG, Self RAG, and Corrective RAG

  • Link: https://inf.run/hTwjC

Recommended for
these people!

Who is this course right for?

  • Those interested in the RAG system using LLM

  • If you want to start an AI project using LangChain

  • Learn how to evaluate RAG search and generation performance

Need to know before starting?

  • Python

  • LangChain Basics for Beginners (Lecture)

Hello
This is 판다스 스튜디오

Students

2,835

Reviews

168

Rating

4.8

Courses

6

안녕하세요. 저는 파이썬을 활용한 데이터 분석 및 인공지능 서비스 개발 실무를 하고 있습니다. 관심 있는 주제를 찾아서 공부하고 그 내용들을 많은 분들과 공유하기 위해 꾸준하게 책을 집필하고 인공지능 강의를 진행해 오고 있습니다.

 

[이력]

현) 핀테크 스타트업 CEO

전) 데이콘 CDO

전) 인덕대학교 컴퓨터소프트웨어학과 겸임교수

Kaggle Competitin Expert, 빅데이터 분석기사

 

[강의]

NCS 등록강사 (인공지능)

SBA 서울경제진흥원 새싹(SeSAC) 캠퍼스 SW 교육 ‘우수 파트너 선정’ (Python을 활용한 AI 모델 개발)

금융보안원, 한국전자정보통신산업진흥회, 한국디스플레이산업협회, 대구디지털산업진흥원 등 강의

서울대, 부산대, 경희대, 한국외대 등 국내 주요 대학 및 국내 기업체 교육 경험

  

[집필]

 

[유튜브] 판다스 스튜디오 : https://youtube.com/@pandas-data-studio?si=XoLVQzJ9mmdFJQHU

Curriculum

All

79 lectures ∙ (8hr 42min)

Lecture resources

are provided.

Published: 
Last updated: 

Reviews

Not enough reviews.
Become the author of a review that helps everyone!