Rollback before you deploy: a Friday checklist
Expand-then-migrate schema changes, feature flags and a 15-minute abort path we rehearse in staging.
Read article →Write-ups from work we've shipped: what we measured, what we changed and what happened next.
Every post includes a concrete technical detail. If we can't show the before and after, we don't publish it.
Expand-then-migrate schema changes, feature flags and a 15-minute abort path we rehearse in staging.
Read article →NAT gateways, snapshots, debug logs and oversized databases — why the invoice jumps after leaving shared hosting.
Read article →If the screens already work, we usually replace the backend, not the UI. When a full rewrite is actually justified.
Read article →Composite keys from slow-query logs, covering indexes, and when one more index makes writes worse.
Read article →Test gates, staged deploys, rollbacks and audit trails — the minimum before a pipeline counts as done.
Read article →Custom slugs, kill switches, expiry and an API — what we built into srt.to, and why.
Read article →Query batching, indexes, Redis and a strangler-fig backend swap, with the same UI for dispatchers.
Read article →Share your email and we'll follow up within one business day with a few questions — then a written estimate.