Blog / Writing
Notes on infrastructure, platform engineering, and the occasional postmortem-shaped lesson.
July 24, 2026
What "cloud-native" actually buys you for an AI platform
Model serving is the easy part. The infrastructure layer underneath — autoscaling, cost controls, and observability — is where a cloud-native AI platform is won or lost.
January 25, 2026
REST, gRPC, GraphQL: When to Use Each (And Why Teams Pick Wrong)
Not all APIs are built equal. REST is for flexibility, gRPC for speed, GraphQL for client complexity. When to pick which—and what it costs you.
January 20, 2026
20 Years of Cloud: From EC2 to the Infrastructure Revolution
AWS 2006 to 2026—how cloud went from radical to table-stakes to invisible. The innovations that changed infrastructure forever, in one concise journey.
January 15, 2026
Running Local LLMs: How Small Enterprises Win Without the Cloud Bill
Skip the API calls, keep the data private, run smaller models locally—and your infrastructure cost stays closer to zero. A practical guide to serving LLMs on hardware you already own.