AWS · MLA-C02 · BETA
Embeddings, vector stores, and RAG documents
MLA-C02 · Lesson 4 of 23 · 24:50
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 4 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 1, data preparation). In this lesson: embedding models and their dimensions, vector stores (OpenSearch, pgvector, S3 Vectors), data types and text pre-processing, and chunking and metadata for retrieval augmented generation. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- An internal tool must search fifty million document chunks. It runs a few queries per hour, an answer under one second is fine, and storage cost matters most. Which vector store fits best?
- A text classifier routes support tickets, but tickets written with local slang for the service are often misrouted, and the training data has few such examples. What should the ML engineer do?
- Customers ask detailed questions about single clauses of long rental contracts. Small chunks find the right clause, but answers miss the conditions around it. Which chunking strategy should the ML engineer choose?