Data Engineer Job Description Template

Production pipeline design · 24%
This role owns well-defined scopes; success depends on turning requirements into reliable pipeline architecture. Evaluated in Round 1.

• Weak: Builds pipelines without clear handling for failure modes or operational concerns
• Meets the bar: Designs pipelines with appropriate staging, retries, and maintainable structure
• Strong: Designs end-to-end pipeline architecture with robust error handling, idempotency, and clear data flow guarantees

SQL and data modeling · 19%
Strong SQL and modeling are required to transform data accurately and support downstream consumption. Evaluated in Round 1.

• Weak: Uses SQL that is difficult to validate or maintain; modeling choices lead to frequent inconsistencies
• Meets the bar: Writes correct, maintainable SQL and uses modeling patterns that match requirements
• Strong: Designs modeling that prevents anomalies (e.g., duplicates/partial updates) and enables efficient, correct analytics queries

Data quality and validation · 19%
Preventing bad data and detecting issues early directly impacts stakeholder trust and reduces incidents. Evaluated in Round 2.

• Weak: Adds limited checks; failures are discovered only downstream
• Meets the bar: Implements validation rules and clear failure signaling tied to pipeline stages
• Strong: Implements layered quality gates and evidence-based diagnostics that speed up root-cause analysis

Workflow orchestration · 17%
Orchestrating production pipelines is central to reliability, scheduling, and operational control. Evaluated in Round 2.

• Weak: Relies on manual intervention for schedule/ordering or lacks control for dependencies
• Meets the bar: Configures orchestration with retries, dependencies, and environment separation
• Strong: Implements orchestration patterns that ensure deterministic runs (e.g., backfills) and consistent behavior across environments

Debugging and incident response · 15%
Mid-level engineers must independently debug and resolve production data issues. Evaluated in Round 3.

• Weak: Troubleshoots slowly or makes changes without isolating the cause
• Meets the bar: Diagnoses failures using logs/metrics, validates hypotheses, and applies corrective actions
• Strong: Leads fast root-cause analysis, documents learnings, and prevents recurrence with targeted safeguards

Cloud production engineering · 6%
Cloud deployment knowledge is needed to operate pipelines reliably in the production environment. Evaluated in Round 3.

• Weak: Limited understanding of cloud runtime constraints or deployment practices
• Meets the bar: Deploys and operates pipelines using appropriate cloud services and permissions
• Strong: Designs for reliability and security in cloud environments (least privilege, environment controls, reproducible deployments)