From the course: AI Engineering Use Cases and Projects on AWS: Production-Grade LLM Systems

Unlock this course with a free trial

Join today to access over 25,300 courses taught by industry experts.

AI-engineering capstone

AI-engineering capstone

From the course: AI Engineering Use Cases and Projects on AWS: Production-Grade LLM Systems

Start my 1-month free trial Buy for my team

AI-engineering capstone

“

Here we have an AWS Lambda architecture that uses Bedrock. So what's really interesting about this is you have a global-scale, serverless, very efficient way to use large language models because you can talk to the foundation model interface, you can put in extremely energy-efficient resources. So a lot of people are talking about the paradox where as you get more efficient, you use more. But if we think about using high-level interfaces, like for example, serverless with Bedrock foundation interface, and you even put your own tiny models or your distilled models on top of there or even use these open source models like DeepSeek, this is a great way to take advantage of those capabilities but also be energy efficient. So the idea here is you have an API gateway. You talk to this Rust Lambda. It has massive, incredible performance, basically as good as you can get. And you have adjacent requests, and you have model invocation. Some of the characteristics include cold start. So 50…

Contents