Dev Summit 24 Front of the frontend Session 2

The whole world is excited about generative AI, but how do we start to build with it? Do we need to learn linear algebra, machine learning, or even python?

It turns out that our existing knowledge and skills are still very much in demand. There are some terms and tools you need to understand, but it's not as big a jump as you think.

This talk is a roadmap for understanding GenAI as a developer and how to start building with it, from interacting with large language models to rendering output to the browser and everything you need to know in between.

In this talk we’ll break down keyword, semantic, vector and hybrid search approaches, exploring how each method works, their advantages and disadvantages, and practical use cases.

The talk will break down what can be overly complex concepts into practical takeaways for our everyday work as JavaScript developers. By the end of the session you’ll have a better understanding of when and how to use different search techniques to optimize your user experience. Most importantly you will also learn how as a JavaScript developer we use can use these search techniques using open source tooling and libraries including some powerful search engines written in TS/JS.

We started a brand new AI team with 14 engineers that have never touched AI before! We started madly learning about AI, LLMs, top-k, temperature, ALL THESE NEW WORDS!

After 6 months, we’ve learned so much, we’re productive and it feels like we’re building something special.

I’ll describe how we quickly became productive, tips to give a large team focus, methodical thinking and some examples how we’re really all just problem solvers.

The audience will get a taste of what it’s like to start prompt engineering for the first time, how backwards it feels and how we can do better.