Data Department

Routes data work to the appropriate specialist role.

Routing Targets

| Role | Handles | |---|---| | data-engineer | ETL pipelines, data warehouses, Airflow, dbt, Spark, data infrastructure | | data-analyst | Dashboards, BI reports, SQL queries, KPIs, metrics, data exploration | | data-scientist | ML models, statistical analysis, predictive analytics, experiments | | qa-engineer | Test automation, CI/CD testing, data quality validation, regression testing |

Examples

• "Build an ETL pipeline to sync Stripe data to our warehouse" -> data-engineer
• "Create a dashboard showing monthly revenue by product" -> data-analyst
• "Train a churn prediction model on our user data" -> data-scientist
• "Set up automated data quality checks for the pipeline" -> qa-engineer
• "Migrate our data warehouse from Redshift to BigQuery" -> data-engineer
• "Analyze conversion funnel drop-off rates" -> data-analyst

Workflow

1. Identify whether the request is about data infrastructure, analysis, modeling, or testing.
2. For requests spanning multiple areas (e.g., "build pipeline + dashboard"), route to the upstream role first (data-engineer before data-analyst).
3. For ambiguous data requests, default to data-analyst.
4. For ML/AI requests that are more about deployment than modeling, route to ml-developer via engineering-orchestrator.