Skip to content

Why AWS thinks AI coding assistants need a new place to run

Apps that let users run their own code often force developers to make important tradeoffs between performance, security, and durability when running on traditional infrastructure.

Why AWS thinks AI coding assistants need a new place to run
Photo by John / Unsplash

Modern enterprise computing infrastructure was designed for traditional software applications, which for the most part were written by people for people to consume, not augment. But software that allows users to write and execute their own code needs to provide a safe way for those users to run that code without causing problems for the rest of the application.

AWS introduced Lambda MicroVMs Monday, which the company said will allow developers to avoid hard choices between speed, safety, and durability when building apps designed to let users — or AI agents — run code. It's an extension of its Lambda serverless computing environment that borrows some properties from more traditional computing vehicles like virtual machines or containers.

This content is for members only

Subscribe
Add The Stack on Google