Intermediate

Introduction to RAG

A practical, intermediate deep dive into retrieval-augmented generation — how embeddings and vector search work, how to chunk and index documents, how reranking sharpens results, and how to measure whether a search system is actually good.

From Keyword Matching to Meaning

1
Why Search Needed to Change3 min read
2
Three Ways Language Models Power Search3 min read
3
How Embeddings Capture Meaning3 min read

Building the RAG Pipeline

4
The RAG Pipeline, Step by Step4 min read
5
Loading, Splitting, and Storing Documents4 min read

Chunking Strategies

6
Why Chunk Size Is a Design Decision3 min read
7
Fixed-Size, Token, and Overlapping Chunks3 min read
8
Semantic Chunking3 min read
9
Agentic Chunking4 min read

Ranking, Storing, and Measuring Results

10
Similarity and Distance Metrics3 min read
11
Choosing a Vector Database3 min read
12
Reranking: A Second Pass for Relevance3 min read
13
Evaluating Search Quality with MAP4 min read