Hire data engineers who build pipelines that fail loudly
Centricone places data engineers screened on the things that decide whether a pipeline is trustworthy: idempotency, late-arriving data, and what happens when a source changes shape without warning.
Screening
What we check that a keyword match does not
Every candidate sits a live assessment with an engineer who does this work. For this role, that conversation covers:
- Idempotency and reruns — whether a failed job can simply be run again
- Handling of late, duplicate, and malformed records rather than the happy path
- Warehouse modelling: how they structure for the questions the business asks
- Observability of data itself, not only of the job that moved it
What they do once they are in your team
Building and maintaining ingestion and transformation pipelines
Warehouse modelling in Snowflake, BigQuery, or Postgres
dbt models, tests, and documentation
Data quality monitoring and alerting on the metrics the business relies on
Which level you actually need
Most briefs ask for senior by default. Sometimes that is right; often the work is better served by a mid-level engineer and a clearer specification.
Builds and maintains pipelines within an established platform and modelling convention.
Designs the pipeline and warehouse architecture and owns data quality outcomes.
Owns the data platform, its cost, and the standards other teams build against.
Questions about hiring data engineers
Not sure what you need? Describe the work and we’ll tell you which role fills it.