Must have:
At least 3–5 years in a data engineering role or something close to it, building pipelines other people depend on
Hands-on data science experience. You do not need a research background. You do need to have built and evaluated models yourself: enough to tell a real anomaly from noise, to judge when a finding is solid enough to put in front of a paying customer, and to know what a pipeline owes a model running in production
Solid with Python and SQL and cloud data infrastructure. You have designed and run pipelines, APIs and workflows in production
Comfortable with high-volume time-series data: ingestion, data quality, backfills, and the everyday reality of late, missing and duplicated readings
You think end to end before you go deep. When you hit a hard problem, the first instinct we want is “can a change somewhere else in the chain make this easy?” rather than heading straight into solving the hard version. A company our size cannot afford clever answers to problems that did not need solving
Clear written and spoken English. German is a plus
Nice to have:
Energy, industrial, IoT or another domain where the data comes off real hardware
Metering or device protocols, telemetry at scale, or streaming pipelines
Knowledge of a second programming language, like TypeScript or Kotlin
You run what you build: monitoring, alerting, and the occasional early morning debugging a pipeline
Copyright © 2026 Grabjobs Pte.Ltd. All Rights Reserved.