Kafka consumer lag is not a metric — it is a symptom
How we cut lag from 4.2M messages to near-zero by fixing partition keys before buying bigger brokers.
Read articleField notes from production
Everything here came out of an incident, a migration or a Tuesday afternoon spent fixing something that should never have broken. No thought leadership — just what worked.
Latest
Most read this month
6 articles from the engineering team.
Kafka
Redis
The cache layer nobody measured
Hit ratios, stampede protection and the three-line change that took our p95 from 900ms to 120ms.
Read articleCI/CD
Our CI/CD checklist for teams of eleven
Preview environments, secret scanning and blue-green releases — the pipeline we reuse on every engagement.
Read articleFull archive
Every article
Jump straight to one.
The cache layer nobody measured
Hit ratios, stampede protection and the three-line change that took our p95 from 900ms to 120ms.
Read articleOur CI/CD checklist for teams of eleven
Preview environments, secret scanning and blue-green releases — the pipeline we reuse on every engagement.
Read articleHardening a Linux VPS in ninety minutes
The exact checklist we run on every client server — SSH, firewall, TLS, backups and monitoring.
Read articleSelf-host n8n or pay Zapier forever?
A break-even calculation from real usage numbers, and the migration steps if you go self-hosted.
Read articleWe made WordPress fast again
LCP from 4.8s to 1.1s on a 12-year-old WooCommerce install — without a rebuild, without losing content.
Read articleStay close
One engineering note a month
No newsletter theatre. One write-up, one incident write-up, and anything we shipped — straight to your inbox.