DevOps people have always lived between worlds. One day the problem is a slow deployment. The next day it is a Kubernetes networking issue, a Terraform drift, a flaky test suite, an unexpected cloud bill, a security exception, or a production incident that refuses to fit neatly into anyone’s dashboard. That variety used to make DevOps look like a discipline of generalists. In the AI era, that is not a weakness. It is the point. ...
I’ve been running mixed clusters with ML training jobs and regular services for about two years. Scheduling has been the biggest headache. A distributed training run would get only some pods placed, GPUs would sit there doing nothing, and everyone would lose time. Kubernetes 1.35 came out last week, so I spent the weekend testing it on our staging cluster. A few of these changes are genuinely useful. Gang Scheduling Finally Exists The biggest addition is workload-aware scheduling with gang scheduling support. It’s still alpha, so I would not put it in production yet, but the model is exactly what we needed: a group of pods either gets scheduled together, or not at all. ...
🤖 A New Era of Creation We’re living through something extraordinary. AI isn’t just automating tasks, it’s amplifying imagination. As someone who’s spent years deep in the world of infrastructure, automation, systems, and DevOps, I find myself more excited than ever. Not because AI replaces what we know, but because it lets us build upon it faster than ever before. 💡 Ideas Used to Be Worthless Without Execution For most of IT history, ideas were… expensive. You could have a vision, a product concept, a technical improvement, but without: ...