Most recent · Orchestration
Oozie → Airflow: what does not translate
An Oozie coordinator waits for data to exist. An Airflow DAG fires on a clock. The whole difficulty of an orchestration migration sits in that gap.
Writing
What I could not find written anywhere
Eight pieces that came out of questions an engagement forced me to settle — orchestration, Spark migrations, data quality, risk engines. None of this is a rewritten documentation page.
All writingRSS ↗
08.2026Oozie → Airflow: what does not translateOrchestration
07.2026Spark 3 and old dates: the proleptic calendar trapSpark
07.2026A row set aside in a regulatory filing is an under-declarationData quality
06.2026AQE, DPP, skew join: what Spark 3 really fixesSpark
06.2026Java 8 → 17 under Spark: the list of surprisesPlatform
05.2026Introducing TDD in a team that has never done itMethod
05.2026VaR: the shape of the computation comes before the optimisationMarket finance
04.2026Building a payments lakehouse alone: what it forces you to settleArchitecture
One essay a month, nothing else
No promotion. One click to unsubscribe, and the archive stays open without an address.