Change Version

FROM THE FIELD

Stuff we've learned the hard way

15 years of production infrastructure. Some patterns, some decisions, some things that went wrong.

OBSERVABILITY

The 3-tier observability setup that works for 90% of startups

Most startups either monitor nothing or monitor everything. Both are wrong. Here's what we set up by default and why.

8 MIN READ
ARCHITECTURE

Why we stopped using service mesh for teams under 20 engineers

Istio is powerful. It's also a full-time job. For most teams, it's complexity you don't need yet.

6 MIN READ
NETWORKING

Our default VPC layout and why it looks like that

Public subnet, private subnet, NAT gateway, security groups. The boring foundation that prevents exciting outages.

5 MIN READ
ARCHITECTURE

When to use ECS instead of Kubernetes

More often than you think. If your team is under 10 and you're on AWS, K8s might be overkill. Here's how we decide.

7 MIN READ
CI/CD

The CI/CD pipeline structure we copy to every project

Lint, test, build, scan, deploy. Five stages, no surprises. We've refined this over 50+ projects.

5 MIN READ

This is the kind of experience that doesn't come from tutorials. If you want someone on your team who's already made these mistakes, let's talk.

Want to see if we're a fit?

30-minute call. We'll ask about your stack, you'll ask about us. No pitch deck, no pressure.