Retrieval-Augmented Generation from Scratch
Build a RAG system that actually retrieves the right thing: embeddings, chunking, hybrid search, reranking, and grounded answers.
COURSE · AI ENGINEERING
Retrieval-Augmented Generation from Scratch
12 LESSONS
YOUR INSTRUCTOR
Priya Raman
Priya builds products on top of language models and, more importantly, the evaluation harnesses that tell you whether they actually work. She has shipped retrieval systems over messy, real-world corpora. Her position is that almost every bad AI feature is a retrieval or evaluation problem wearing a prompt-engineering costume.
View instructorWhat you'll learn
Understand embeddings
Know what a vector encodes, what it does not, and which similarity measure to use.
Chunk content well
Split documents so retrieved passages are self-contained and carry useful metadata.
Improve retrieval quality
Combine keyword and vector search, then rerank, and measure whether it helped.
Ground every answer
Cite sources, and make the system say it does not know instead of inventing one.
Course content
4 MODULES · 12 LESSONS · 1H 57M TOTAL
MODULE 01 Embeddings and Vector Search
The retrieval primitive: turning text into vectors and finding the near ones quickly.
3 LESSONS · 21M
MODULE 02 Ingestion and Indexing
Getting your content into the index in a shape that retrieves well and stays current.
3 LESSONS · 32M
Every lesson, searchable to the second.
© 2026 BIBLION
PRIVACY · TERMS · CONTACT