AWS · MLA-C02 · BETA
Foundation models, customization, and RAG patterns
MLA-C02 · Lesson 8 of 23 · 1:01:55
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 8 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 2, model development). In this lesson: choosing a foundation model in Amazon Bedrock, customization strategies (prompting, RAG, fine-tuning, distillation, LoRA), RAG patterns and retrieval tuning, and latency and cost tradeoffs. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- Prices and fees change every month in each city, and the assistant must quote the current values, with their source. The team wants to avoid retraining. What should the ML engineer use?
- Customers often ask about a fee by its code, like F12, and the vector search returns similar fees, but not the exact one. What should the ML engineer do first?
- Every request starts with the same block of rental terms, 6,000 tokens long, then a short question, and costs and latency are too high. What should the ML engineer do?