I Made a Million-Dollar Mistake, Now What?

We’ve all shipped something that has cost a business money. Forgetting an expensive AWS service running all weekend, pushing test keys to prod, misconfiguring a webhook, it happens. But what does that look like when you mess up with tens of millions of users, and over time rack up hundreds of thousands in losses? I once shipped something that ended up costing the business a stupid amount of money. Not because I was careless, but because the system looked “fine” until it very much wasn’t. This talk is about what happens after that moment, when the dashboards stop making sense, the numbers don’t line up, and you are left trying to explain a loss that doesn’t even have a clean shape yet. This is not a hero story or a blame story. It’s a talk about how you deal with business risk when you’re running small, lean teams in startup or scale-up mode, with messy processes, constant urgency, and very little time to “do things properly”. It’s a retrospective on engineering practices and how they map, or fail to map, to reality, and how you actually move the needle outside the scope of commits and lines of code.