Build a Payroll Revision Attribution Desk

A labor economist reviewing a monthly employment release.

Separate new payroll growth from revisions to earlier months. Give an economist an auditable release-to-release account of why the labor-market narrative changed.

Key: reference_month + release_id + series_id.

Fields: employment_level, unit, seasonal_adjustment, benchmark_version, release_time.

Compare identical reference months. Do not combine levels from incompatible benchmark bases.

Align reference months across two releases.

Calculate prior-month revisions separately from new-month growth.

Flag benchmark or definition changes.

Export a release bridge and annotated research memo.

A release comparison desk with revision bars, level-to-growth drilldown, and benchmark alerts.

An archived-release alignment pipeline.

A revision bridge with source-linked narrative sentences.

Runnable local application, setup commands, and a dependency lockfile.

Versioned input data, data dictionary, acquisition instructions, and rights notes.

Calculation traces, source references, assumption records, and exported review artifacts.

A deterministic baseline and tests for normal, incomplete, contradictory, and failure-recovery cases.

A recorded complete workflow with a reproducible manifest and explicit limitations.

Repeated ingestion produces the same release comparison.

Revisions use matched reference months only.

Benchmark changes cannot silently enter a comparable-period total.

The memo distinguishes new growth from revisions numerically.

Change only the revised history. The new-month estimate remains unchanged while the revision total moves.

The attached starter files are prototype inputs. Expected starter answers are visible. They are not independent holdouts.

Full target: Acquire 36 archived employment releases with monthly levels and revision notes. Preserve benchmark changes as separate events rather than ordinary new observations.

Complete target contracts and expert review remain pending. Public observations are not historical release vintages.

DuckDB, pandas, Altair

Use an available model appropriate to the task. Record provider, model identifier, version or retrieval date, settings, cost, and latency. Use deterministic code for calculations. Models may extract, retrieve, or explain evidence. They must not invent missing records, override approval rules, or claim professional authority.

A revision is not evidence that an earlier release was deceptive.

No fabricated records, source citations, professional endorsements, or expert-adjudication claims.

No future-data leakage into dated decisions, model fitting, or historical valuation inputs.

No unauthorized payments, account restrictions, regulatory filings, or production changes.

No automated appraisal certification, tax-compliance determination, or unsupported fraud accusation.

No concealed uncertainty or silently invented inputs required for a numerical conclusion.

Correct domain calculations and justified methods with documented assumptions. Weight: 30.

Source lineage, temporal integrity, rights checks, and explicit synthetic-data labels. Weight: 25.

A working application with review, export, error handling, and reproducible state. Weight: 20.

Adversarial cases, uncertainty, missing-data behavior, and sensitivity analysis. Weight: 15.

Evidence-linked conclusions that distinguish facts, assumptions, and unresolved judgments. Weight: 10. Score each dimension from zero to two. Weighted score = sum(weight * score / 2). A failed safety gate blocks acceptance.