lessons from Astro

There rarely comes along an opportunity as exciting as getting to push the web forward. Even rarer is an opportunity that knocks on your door. But such was my luck when Fred Schott gave me a phonecall and asked if I wanted to help him start Astro (nÊe Skypack).


A growing frustration with frontend build tools had led me to experiment with ESM-focused bundlers. This led me to work on Snowpack early on, and I was the first core contributor to the project other than Fred. I had worked on a few features off-and-on for the better part of a year when Fred called in May 2020 offering me a fulltime job. I accepted, of course.

In the months that followed, we launched:

  • ESNEXT CONF, a worldwide conference discussing the future of ESM / JavaScript
  • Several major versions to Snowpack, inspiring the next generation of build tools like Vite
  • Skypack, a powerful CDN that let users use almost any package from npm without Babel or build tools
  • Astro, a game-changing site build tool

And all the while, Fred and I discussed business strategies, ways to change the internet, ways to improve the web scene. And Astro was what we came up with (by then, with input from the rest of the team we hired).