Latest articles
How we route traffic across 30 regions with anycast
A deep dive into the network architecture that powers sub-20ms latency for users on every continent.
Secrets management and credential rotation across a fleet
Long-lived API keys are the quiet liability in most infrastructures. Here is how we moved every service to short-lived, automat…
Structured logging that stays affordable at scale
Logs are the first thing you reach for in an incident and the first bill that spirals out of control. We rebuilt our logging pi…
A multi-region failover runbook you can actually trust
A failover plan you have never rehearsed is a hope, not a plan. Here is the runbook we test every month so a region loss is bor…
Right-sizing compute: cutting cloud spend without downtime
Most clusters are paying for headroom they never use. We trimmed compute spend substantially while improving tail latency — her…
Feature flags: shipping safely behind toggles
Deploying code and releasing a feature should be two separate events. A flag system lets you ship continuously and release deli…
Distributed tracing: from a latency spike to the root cause
When a request is slow, the question is always "where?". Tracing turns that guess into a timeline you can read.
Background jobs and queues that survive a bad day
The work that happens after the response is sent is where reliability quietly breaks. Here is how we made our async tier boring.
From git push to global deploy in 90 seconds
A fast pipeline changes how a team works. When deploys are quick and safe, people ship small and often. Here is how ours runs.
Chaos engineering: breaking things on purpose
You do not really know your system is resilient until you have watched it fail on your own terms. Chaos experiments turn assump…
Point-in-time recovery: backups you can bet on
A backup you have never restored is just a rumour. Point-in-time recovery turns "we think we can recover" into a number you can…