Queries Josh Lane's org headcount from a DuckDB at ~/workspace/areas/staffing/staffing.duckdb.
Source: HRIS "Opex TD" export, company code 80405 (Engineering, CS, CX, Professional Services, Product, UX, WeSupply, post-sales). Does NOT include G&A, Finance, Sales, Marketing, Legal — use BigQuery dim.users for company-wide data.
Current snapshot: 2026-05-18 — 245 active, 233 inactive/past
Execution
For predefined queries, use just from within ~/workspace/areas/staffing/. For custom queries, use the DuckDB CLI directly:
duckdb ~/workspace/areas/staffing/staffing.duckdb "SELECT ..."
See references/query-patterns.md for copy-paste SQL for all common scenarios.
Views
current_headcount— active employees only from the latest snapshotcurrent_all— active + terminated from the latest snapshotheadcount_history— all snapshots; useWHERE snapshot_date = 'YYYY-MM-DD'to pin to a specific date
Key Gotchas
Supervisor lookup for Josh: supervisor = 'Lane, Joshua' (not "Lane, Josh" — the just reports recipe is broken for this case). Use WHERE supervisor LIKE 'Lane%'.
Customer Support is fragmented: Four separate CC1 values — aggregate with WHERE cc1_description LIKE 'Customer Support%'.
Engineering is fragmented: Multiple CC1 values — aggregate with WHERE cc1_description LIKE 'Engineering%'.
work_country = 'HOME' means domestic/US remote (85% of headcount). International employees: IND=24, ROU=7 (WeSupply Romania), CAN=2, BRA/IRL/PHL=1 each.
annual_salary: ~15% NULL for active employees. Always filter WHERE annual_salary IS NOT NULL before averaging.
Full Schema
See references/schema.md for the complete column catalog, CC1 values, and work_country reference table.