Kiril Peyanski

AI is changing how we build UIs. This talk shows why verbose, composable APIs work better for humans and machines, and what the next generation of UI libraries looks like. We will contrast old shortcut-driven APIs with modern patterns that give both developers and AI assistants the context they need to build faster and better.

Every modern app starts from the same truth: your website is built on data. Traditionally, developers define how that data turns into UI - the f in UI = f(data, state). But every user approaches your app with different goals, contexts, and focus, and a one-size-fits-all interface forces them to work around the UI instead of it working for them. In this talk, we’ll explore the Generative Front-End - a new paradigm where the logic that maps data to interface is itself generated by an LLM. We’ll see how React Server Components and Server Functions make this pattern possible natively, without hacks or ad-hoc APIs. The result: interfaces that adapt to the user, the data, and the moment - where developers design the system that builds the screens.

React Server Components are changing how we think about component architecture-but in practice, most real-world UIs don’t live purely on the server or the client. In this talk, we’ll explore how to build hybrid components that span both, maintaining a clean separation of concerns while delivering interactivity and performance.

Explore Server Components through the lens of reusable UI Component, where everything "depends" on the individual requirements of the use-case and individual application needs. Instead of fighting over 'server' vs 'client' - let's have the best of both worlds.