Stripe: How We Rapidly Train Machine Learning Models with Kubernetes

This post is...intense. It's one of the deepest "behind-the-scenes here's how our ML infrastructure works" posts, and as such it's quite notable. The infrastructure described in this post is impressive, and far beyond what most teams have access to.

I recently spent a bunch of time with a good friend who works in data, and we spent a lot of time talking about the pluses and minuses of different jobs in the field. It really made me recognize just how important tooling is as a part of the vetting process for a new job in data: employees at companies with advanced tooling are just far more effective than employees at companies with no or poor tooling. That means they do more valuable work, get better experience, and level up faster. They also just tend to be happier.

In your next interview process, make sure you learn about the team you'll be working on and the tooling you'll have access to. Weight that heavily in your decision criteria.


Want to receive more content like this in your inbox?