Movies have promised the future where we can talk to AI Assistants with natural language and have them respond with all the nuances of humans. However for years we've been stuck with really "stupid" AI experiences over voice. With the advent of LLMS this is finally a reality!

Lots of different technologies have come together to create amazing audio experiences. However, there are some technical problems that we need to solve. In this talk we will learn how to chain these technologies, solve some of the common problems and build a new JARVIS together, powered by the LLM of your choice.

If you write JS or TS, probably every tool you use daily is currently being rewritten in Rust.

This has advantages (have you heard that Rust is "blazing fast"?), but also potential downsides.

The evolution of JS from the dark days before async/await was fuelled by tooling written in JS. If our tools migrate to a language we don't know, will we lose the ability to control or customize them?

In this talk, I'll explain how in Oxc (the compiler underlying the next generation of the open source Vite bundler), we're trying to "square the circle" - providing rich customisation via plugins written in JS, but maintaining the exceptional performance of Rust - and take you into the nerdy details of memory layouts, allocators, garbage collection, and the tricks that make it possible to have our Rusty cake and eat it with a JS fork.

This talk introduces a crucial but often overlooked aspect of AI application development: evaluation-driven development (EDD). Using a simple LangGraph agent as a practical example, we'll demonstrate why and how to build a robust evaluation framework that goes beyond simple unit tests. We'll explore the importance of continuous evaluation during the development cycle and how this practice directly translates to the need for comprehensive observability in production, ensuring your AI application remains accurate, reliable, and effective in the real world.

Remember when profiling Node.js meant wrestling with perf, sed scripts, and mysterious V8 internals? Traditional flame graph generation has been like performing surgery with a rusty spoon, requiring kernel-level tools and arcane command-line incantations. The barrier to entry is sky high. The time to insight is glacial.
We've completely reimagined this at Platformatic with @platformatic/flame - a zero-config profiling tool that works instantly. Just run flame run server.js and get production-grade flame graphs. Combined with our WebGL-accelerated React component library (react-pprof), we handle massive datasets with buttery-smooth 60fps interactions. This isn't just another tool - it's about democratizing performance analysis.
When your Auth Service hits 96% Event Loop Utilization, you need answers NOW. I'll show you our ecosystem-compatible architecture, how we profile microservices in-memory with Watt, and real production cases where this saved millions in infrastructure costs. Performance is everybody's responsibility, but the tools have been nobody's friend until now.