How We Rebuilt GitLab CI on GKE Spot VMs
How we moved GitLab CI to GKE with shared BuildKit caching, multiple Spot machine types, and an on-demand fallback for scarce capacity.
Posts about devops
Browse 36 posts in this category
Team-Based DLP: Give Each Group Its Own Redaction Rules
One forwarder, many teams, one DLP rule nobody wants to touch. Scoped group rules let each team own its redaction. How-to for proxymock web and Speedscale.
Five Ways to Use OpenTelemetry Beyond Observability
Teams run inventory, security logs, request routing and billing through OpenTelemetry. Five non-observability uses, and the failure mode of each.
Your Observability Stack Found the Fire. Congratulations.
Observability tells you when production is on fire. Capturing and replaying real traffic lets you recreate the incident, verify the fix, and keep it fixed.
Reliability Engineering in the AI Era
See why reliability engineering must span code, testing, telemetry, and incidents as AI agents erase the boundary between pre-production and production.
The Best Model Is a Routing Decision
The best model isn't the smartest — it's whichever should get the next unit of work. Notes from NVIDIA on routing, local models, and the AI factory.
AI writes code in seconds, but delivery still takes days
AI pushed throughput up 59%, yet median delivery got worse. The bottleneck moved to validation. How replaying production traffic in CI closes the gap.
The Three Pillars Were Built for Humans
Metrics, logs, and traces were built for humans and cheap storage. AI inverts both assumptions, and the next maturity level is a deterministic replay sandbox.
Five things your logs will never tell you
Logs, metrics, and traces are a lossy compression of production. Five things you can do with a traffic data lake that observability can't.
Try Speedscale free
Choose the desktop proxymock or the hosted cloud trial to get started.