Big dataPipelines
Pipelines that think (and ones that only pretend)
Building data paths where freshness, correctness, and cost are design inputs — not postmortems.
Continue reading...Staff+ engineer
I write about the work that sits above “make it work”: OpenTelemetry that tells the truth, data platforms that survive load, and AI integrations that behave in production. Depth over demos. Tradeoffs over buzzwords. More about me.
Big dataPipelines
Building data paths where freshness, correctness, and cost are design inputs — not postmortems.
Continue reading...AIProduction
Integrating LLMs and ML services like any other unreliable dependency: contracts, observability, and escape hatches.
Continue reading...GoPerformance
When an overnight ETL kept lighting up the heap monitor — and how profiling beat cleverness.
Continue reading...