Skip to content

Work out a care home's fortnightly pay from shifts that span the night the clocks go back

Exact prompt

# Work out a care home's pay for two weeks of shifts Inputs: `shifts.csv` (shift_id, worker_id, start, end, status), `workers.csv` (worker_id, name, hourly_rate), `pay-rules.json` and `rota-note.txt`. Amounts are in pounds. `pay-rules.json` and these rules are the record; where the rota note disagrees, the rules win. Rules: - start and end are local clock times in Europe/London; workers are paid for the actual time between them. - Shifts whose status is cancelled are not paid. - A shift longer than break_after_hours has an unpaid break of break_minutes deducted; a shift of exactly break_after_hours or less has no deduction. - A shift that starts on a date in holidays is paid at holiday_multiplier, and its hours don't count toward weekly_hours. - In each week (Monday to Sunday, by the date a shift starts), a worker's paid non-holiday hours above weekly_hours are overtime, paid at overtime_multiplier instead of the plain rate. - Write `out/payroll.json`: `{"paid_hours": {"<shift_id>": <hours after any break>, ...}, "overtime_hours": {"<worker_id>": <overtime hours over both weeks>, ...}, "pay": {"<worker_id>": <pay>, ...}, "total": <sum of pay>}`. `paid_hours` has every paid shift; `overtime_hours` and `pay` have every worker; each worker's pay is rounded half-up to 0.01 once, at the end. ## How this run works - Work on your own: nobody will answer questions in this task, and writing to `ask/` counts against the run. - Write only under `out/`. Create `out/` if it is missing. Inputs are synthetic; don't fetch anything from the network. - Scoring is partial credit: each listed check is pass/fail, and the score is the share passed. Numbers are compared to within 0.005 unless the task says otherwise.

Turn two weeks of care-home shifts into paid hours, overtime and pay. One Saturday night shift runs through the night the clocks go back, so it lasts an hour longer than the rota clock shows; six-hour shifts sit exactly on the break threshold, a holiday date pays double outside the weekly limit, and some shifts were cancelled. A rota note gets the hours, break and holiday rules wrong. Original benchmark task written for usecaserank.com (not based on a specific public post). Headless, offline run on a reproducible fixture. Each run's notes say how its cost was measured (estimates are marked).

Success criteria: actual_hours breaks overtime pay

Best model for this task so far: Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) (success, $0.0063, 24s).

Run this on another model

All runs (6)

100% pass rate across 6 eligible runs: 6 pass, 0 partial, 0 fail. Partial and failed runs are listed below with everything else.ⓘ

Runs for this use case
ModelCostTimeStep-insⓘAutonomyOutcomeTrustⓘDateCompare
GGemini 4 Argonmedium · thinking on$0.092m0Fully autonomouspassTrust 80%2026-10-23
GGemini 3.8 Flashmedium · thinking on$0.272m0Fully autonomouspassTrust 80%2026-10-23
OGPT-6.1 Soldefault · thinking off$0.0335s0Fully autonomouspassTrust 80%2026-10-23
AClaude Opus 5.5default · thinking on$0.1323s0Fully autonomouspassTrust 80%2026-10-23
AClaude Haiku 5.5default · thinking on$0.006324s0Fully autonomouspassTrust 80%2026-10-23
AClaude Sonnet 5.5default · thinking on$0.0614s0Fully autonomouspassTrust 80%2026-10-23

Embed this result

A live badge for this use case. It links back here and updates as new runs are published.

UseCaseRank badge for Work out a care home's fortnightly pay from shifts that span the night the clocks go back

Get new runs and ranking changes weekly

Which models got better or cheaper on real tasks like this one.