Video course · 15 chapters · 128 min · certificate
RAG with LangChain: Build an LLM Pipeline in 2 Hours
Build a RAG pipeline with LangChain: what RAG is and why LLMs need it, ingestion and retrieval pipelines, document structure, loading text and PDFs, chunking and metadata, embeddings, vector stores, retrieval, LLM integration, advanced RAG and a modular application.
What you'll learn
- Explain RAG and LLM limitations
- Design ingestion and retrieval pipelines
- Load text and PDF documents with metadata
- Chunk, embed and store vectors
- Retrieve context and integrate an LLM
- Structure an advanced, modular RAG app
Chapters
15 chapters · 128:09-
8:14
01Intro Members
What RAG is
Defining RAG and LLM limitations.
-
8:59
02Architecture Members
The RAG pipeline
How RAG prevents hallucination.
-
4:09
03Architecture Members
Technical steps and plan
Two pipelines: ingestion and retrieval.
-
9:01
04Setup Members
Ingestion theory and setup
Data ingestion concepts and environment.
-
8:02
05Documents Members
Document structure
LangChain Document objects.
-
5:11
06Loading Members
Text loading
Loading text files.
-
6:54
07Loading Members
PDF loading
Loading PDFs with a directory loader.
-
5:35
08Chunking Members
Chunking and metadata
Why and how to chunk documents.
-
5:52
09Vectors Members
Vector database theory
What vector databases do.
-
6:50
10Vectors Members
Implementing embeddings
Creating an embedding model.
-
10:12
11Vectors Members
Vector store implementation
Storing vectors in a vector store.
-
11:23
12Retrieval Members
Retrieval pipeline
Turning queries into embeddings and retrieving context.
-
10:10
13Generation Members
LLM integration
Calling an LLM with retrieved context.
-
6:15
14Advanced Members
Advanced RAG
Enhanced pipeline features.
-
21:22
15Engineering Members
Modular RAG application
Restructuring the pipeline into modules.
Jobs this course suits
Our AI checked this course against 500 jobs; these get the most out of it. Each job links to its learning path.