Developing generative AI solutions
Course Overview
What this course covers
The story: Starting a delivery business with a hired driver: decide what you'll deliver, choose a driver, train them on your routes, test them, put them on the road, and keep improving.
In AI/AWS terms: This course walks through the generative AI application lifecycle: define a use case, select a foundation model, improve its performance, evaluate it, and deploy it. Afterwards you should be able to:
- Choose pre-trained models using the right selection criteria
- Define RAG and its business applications
- Explain the cost trade-offs of different ways to customize an FM
- Describe the role of agents in multi-step tasks
- Evaluate FM performance with the right methods and metrics
For the exam: The lifecycle is use case → select FM → improve → evaluate → deploy.