I’ve built my career fixing the hard bugs, bugs that teams had given up on ever fixing. I specialize in the unreproducible crashes, flaky connections, bizarre error messages, and so many “impossible” behaviors that I’ve lost count. I’d call it my superpower, but I’ll tell you a secret. It’s not an art. It’s not magic or genius. It’s not ChatGPT. And it’s not one more web search. The answer is science. There is something true about your program that you don’t know, and you need to find out what it is. And science is how we do that.

This talk will teach the skills you need to understand your program, and by understanding it, to fix it. Many of the techniques aren’t difficult, but developers forget them. Some require thinking a different way. This talk will teach you how to stop thinking like a programmer for a little while, and start thinking a scientist. When you let go of your desire to fix, and center yourself on understanding, then true debugging can begin.