Build a Point-in-Time Economic Release Terminal

An economist preparing a dated research note.

Reconstruct the economic information available at a chosen decision time. Let analysts compare original releases, later revisions, and the conclusions each version supports.

Key: series_id + observation_period + vintage_id.

Fields: value, unit, seasonal_adjustment, released_at_utc, retrieved_at, source_locator.

Missing values remain null. Never fill from a later vintage.

Store immutable release snapshots with hashes.

Apply the decision-time filter before selecting observations.

Recompute changes using only eligible vintages.

Export a dated note with cell-level evidence links.

A time selector, revision comparison table, blocked-record panel, and research-note export.

Versioned observation store and as-of query API.

A note that separates release surprise from historical revision.

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.

A release one second after cutoff is excluded.

Unknown release time blocks intraday eligibility.

A later revision cannot change a frozen earlier export.

Every displayed number links to its selected vintage.

Keep the observation unchanged and move its publication time across the cutoff. Eligibility must change without changing the value.

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

Full target: Acquire 12 monthly series across ten years and their available vintages. Keep release archives for timestamps absent from the API. Exclude unverified intraday records.

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

DuckDB, Polars, Streamlit

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.

Do not claim that currently available observations were known historically.

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.