You’re happily working on your project, surfing through tasks, and delivering value when your boss tells you to add AI to the code. Welcome to the future!

Now you quickly need to catch up with the latest developments in the world of AI which develops at unprecedented speed (even by JavaScript standards). What are the AI components that you can build into your apps? What problems is it good at right now? What systems do you need to architect if you want to bring your own data into the mix?
Do you need to start doing MLOps in addition to DevOps practices?

This session offers a practical take and an opinionated set of tools and practices for working with applications that include AI components. We’ll look at Google’s technical garden: Gemini and Gemma models, AlloyDB AI vector store for your RAG applications, and the Vertex AI platform.
You’ll learn how to work with the open source models locally, what tasks do they excel at right now, how to organize your tests for non-deterministic AI setups, and what to monitor in production to prevent your AI from getting dumber.

All in all you’ll get a quick-paced overview of how to get started and a fully functional sample project which you can use as a starting point for your exploration of building with AI.