#tool building

We take a look at different challenges and decisions when bundling code for web applications. We look at how these are commonly solved and why we need to rethink them.

Engineering problems often repeat themselves in places you wouldn't expect. Sometimes the best solution has already been invented, in a different corner of the software engineering domain. In this talk, we show how and why we mirrored the TCP/IP network stack to solve a communication problem between different components of a browser extension.

Are you building a chat app, a way to see users’ online status or a real-time collaboration dashboard? All of these use cases have one thing in common: Somehow the user-facing application needs to be informed in real-time about events that happen on the backend of your application.In this talk, we’ll look closely at common approaches like polling, application-level updates and pub-sub systems. We’ll explain the tradeoffs with each approach and elaborate why another approach, called Change Data Capture (CDC), is the most elegant and robust way to achieve this.

This talk will explore more advanced ways to write static analysis rules in ESLint using ESLint's control flow APIs. I will quickly explain what a control flow graph is and how you can use it to find issues in your code. I will show you how to detect when a value is assigned to variable uselessly and other logical problems you can detect using this technique.

Your first JavaScript tool might not be the next Babel or ESLint, but it can be built on them! Let's demystify the secret art of JavaScript tools, how they work, and how to build our own. We'll discover the opportunities in our everyday work to apply these techniques, writing our own ESLint rules to prevent mistakes and code transforms to make breaking changes easy to apply. We’ll walk through the fundamentals of working with an abstract syntax tree, and develop our understanding through a live-code. You will be amazed at what you can build, and together we’ll explore how to get started.

Widgets are custom, interactive objects you place in a Figma or Figjam file to extend functionality and make everything a bit more fun. They are written in a declarative style similar to React components, which gets translated to become a node on the canvas. So can you go the other way, from canvas to code? Yes! We’ll discuss how we used the public Figma plugin API to generate widget code from a design file, and make a working widget together using this.