Skip to content

Bill an upholstery workshop's finished jobs from its job cards, fabric rounding and customer deposits

Exact prompt

# Bill an upholstery workshop's finished jobs from its job cards, fabric rounding and customer deposits Inputs: `upholsterers.csv` (upholsterer_id, name, rate_per_hour), `customers.csv` (customer_id, name), `fabrics.csv` (fabric_code, price_per_metre), `jobs.csv` (job_id, customer_id, upholsterer_id, fabric_code, fabric_m, labour_hours, status) and `deposits.csv` (customer_id, amount). All upholsterers, customers, fabrics and jobs are synthetic; prices are in dollars. These rules are the workshop's; where `workshop-notes.txt` disagrees, the rules win. Rules: - Statuses: `done` jobs are billed. `rework` jobs redo the workshop's own faulty work: they are free, but their labour_hours still count as hours worked. `cancelled` jobs count nowhere. - Fabric: a billed job's fabric is charged by the half metre. Its billed metres are fabric_m rounded up to the next multiple of 0.5 (a fabric_m that is already a multiple of 0.5 stays as it is), times its fabric's price_per_metre. - Labour: a billed job's labour is labour_hours × its upholsterer's rate_per_hour. - Discount: a billed job's charge is its fabric plus its labour, less 10% when that sum is 600.00 or more. Don't round until the output. - Hours: an upholsterer's hours are the labour_hours of their `done` and `rework` jobs. - Balances: a customer's balance is the total charge of their billed jobs minus the total of their rows in deposits.csv. - Write `out/workshop.json`: `{"billed_jobs": [<sorted job_ids>], "fabric_metres_billed": <total billed metres>, "discounted_jobs": [<sorted job_ids>], "billed_total": <dollars>, "takings_by_upholsterer": {"<upholsterer_id>": <dollars>, ...}, "hours_by_upholsterer": {"<upholsterer_id>": <hours>, ...}, "balance_by_customer": {"<customer_id>": <dollars>, ...}}`. - Roll-ups: In takings_by_upholsterer, upholsterers in upholsterers.csv with no billed jobs are included at 0. In hours_by_upholsterer, upholsterers in upholsterers.csv with no done or rework jobs are included at 0. In balance_by_customer, customers in customers.csv with no billed jobs and no deposits are included at 0. ## 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.

Bill a synthetic upholstery workshop's week from its job cards. Done jobs are charged fabric by the half metre (rounded up, so a length already on a half metre stays put) plus labour at each upholsterer's hourly rate, with 10% off a job whose charge reaches 600.00. Rework jobs are free but their hours still count, and cancelled jobs count nowhere. Report the billed jobs, billed fabric metres, discounted jobs and total, takings and hours per upholsterer, and each customer's balance after deposits. The prompt says upholsterers and customers with nothing are listed at 0. A length already on a half metre, a customer with only a rework job and an upholsterer with only a cancelled job are decoys. Workshop notes get the rounding, rework, deposit and roll-up 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: billed_jobs fabric_metres_billed discounted_and_total takings_by_upholsterer hours_by_upholsterer balance_by_customer

Best model for this task so far: Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) (success, $0.0041, 15s).

Run this on another model

All runs (3)

100% pass rate across 3 eligible runs: 3 pass, 0 partial, 0 fail. Partial and failed runs are listed below with everything else.ⓘ

Runs for this use case
ModelCostTimeStep-insⓘAutonomyOutcomeTrustⓘDateCompare
AClaude Opus 5.5default · thinking on$0.1120s0Fully autonomouspassTrust 80%2026-10-26
AClaude Haiku 5.5default · thinking on$0.004115s0Fully autonomouspassTrust 80%2026-10-26
AClaude Sonnet 5.5default · thinking on$0.0612s0Fully autonomouspassTrust 80%2026-10-26

Embed this result

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

UseCaseRank badge for Bill an upholstery workshop's finished jobs from its job cards, fabric rounding and customer deposits

Get new runs and ranking changes weekly

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