← All courses
CLOUD
AWS BEDROCK Interview Questions
AWS BEDROCK Interview Questions
4.6 (1968)13,968 learners20 lessons39m
Curriculum
Topic
- 01 - What is Amazon Bedrock2:07
- 02 - How is Bedrock different from Amazon SageMaker1:53
- 03 - What is a foundation model (FM)1:49
- 04 - What are the main features Bedrock offers beyond raw model access1:52
- 05 - Why is Bedrock described as serverless, and what does that mean for developers1:53
- 06 - What is the difference between the Invoke Model API and the Converse API1:53
- 07 - What providers and model types are available in Bedrock1:51
- 08 - How do you choose between foundation models for a given use case (e.g., Titan, Claude, Llama)1:57
- 09 - Explain on-demand, provisioned throughput, and batch inference. When would you choose each1:54
- 10 - Why must fine-tuned custom models use provisioned throughput1:48
- 11 - What is prompt caching and why does it matter for cost1:50
- 12 - What are tokens, and why does token count drive both cost and behavior1:53
- 13 - What is a context window, and what failure happens when you exceed it2:02
- 14 - What is the difference between prompt engineering, RAG, and fine-tuning as ways to adapt a model2:05
- 15 - You're getting Throttling Exception errors at peak traffic. How do you diagnose and resolve it1:56
- 16 - What is Retrieval-Augmented Generation (RAG) and why use it1:42
- 17 - What are Bedrock Knowledge Bases1:49
- 18 - What is the Retrieve and Generate API, and how does it simplify RAG2:14
- 19 - What is chunking, and why does chunk strategy matter for RAG quality2:00
- 20 - What role do embeddings and a vector store play in a Knowledge Base2:21