Practical articles on AI, DevOps, Cloud, Linux, and infrastructure engineering.
We use serverless for specific patterns, not as a default. The patterns where it shines, the ones it doesn't, and the gotchas at production scale.
Building visibility into cloud costs that actually drives action. The dashboards we look at, the alerts that fire, and the queries we run.
How a small team moved from single-region risk to a simple active/passive multi-region setup without doubling complexity.
We run our app in two AWS regions for failover. The hard parts aren't the deployment — they're data consistency, traffic shifting, and the assumptions that break when "primary" is suddenly the wrong region.
We run ~200 Lambda functions. Cold starts, memory tuning, and the cost-vs-latency trade-offs that actually move the bill.
Design for region failure. Active/passive and active/active, data replication, and failover testing.
Design serverless apps for reliability, cold start, and cost. Event-driven patterns and observability.
