Computer Work Val v1.0

An environment data stream and validation set of 300 worlds — drawn from the real work of people earning over $100k a year.

The contributors come from the US occupations with the largest share of labor costs in digital work. Each one turns a piece of their own job into an environment. Computer Work Val v1.0 is 300 of those tasks, chosen to represent every domain, and the environments can be used for both testing and training.

Task authors rate what each failed verifier would cost if it happened at work. Each bar splits a model’s rated failures on one scale.

The most damage the failure could do at work, from an annoyance, to a real cost, to a severe outcome.

How hard the failure is to reverse once it has happened: minutes, real effort, or not at all.

How far the damage reaches: only the person who did the task, their team, or beyond it: a client, a regulator, or the company’s money.

How long the failure would go unseen: right away, at the next review, or late or never.

We build each environment from high-fidelity clones of tools like Jira, Google Sheets and Salesforce, seeded with mock data from the person who actually does the task.

excel.cloud.microsoft

AutoSaveRidgeline_Q3_P&L.xlsx

FileHomeInsertFormulasData

C2fx=SUM('Stripe export'!F:F)

ABC1RidgelineQ2Q32Revenue171,300184,9203COGS(58,900)(61,300)4OpEx(75,200)(78,450)5Net income37,20045,170

P&LStripe export+

Gmail

mail.google.com

GmailQ2 P&L

From ▾Any time ▾Has attachment

DanaQ2 P&L, finalJul 8

XQ2_P&L.xlsx

MarcoRe: Q2 P&L draftJul 2

Stripe

dashboard.stripe.com

RRidgelineSearch

Payments Currency: All

AmountConverted

C$1,250.00CAD$912.50 USD

$640.00USD$640.00 USD

C$3,400.00CAD$2,482.00 USD

$1,180.00USD$1,180.00 USD

Mercury

app.mercury.com

Ridgeline CoffeeChecking ••4021

Balance$212,480.16

To / FromCategoryAmount

SStripeSales+$41,200.00

RRoastworksCOGS−$12,400.00

GGustoPayroll−$2,140.00

environment

The Plato environment

Each tool becomes a sim seeded with their data, with names and accounts swapped for mock ones. Verifiers grade the agent’s work.

Excel

excel.web.plato.sosim

AutoSaveJuniper_Q3_P&L.xlsx

FileHomeInsertFormulasData

C2fx=SUM('Stripe export'!F:F)

ABC1JuniperQ2Q32Revenue171,300188,1003COGS(58,900)(61,300)4OpEx(75,200)(78,450)5Net income37,20048,350

P&LStripe export+

Gmail

gmail.web.plato.sosim

GmailQ2 P&L

From ▾Any time ▾Has attachment

ElenaQ2 P&L, finalJul 8

XQ2_P&L.xlsx

TheoRe: Q2 P&L draftJul 2

Stripe

stripe.web.plato.sosim

JJuniperSearch

Payments Currency: All

AmountConverted

C$1,250.00CAD$912.50 USD

$640.00USD$640.00 USD

C$3,400.00CAD$2,482.00 USD

$1,180.00USD$1,180.00 USD

Mercury

mercury.web.plato.sosim

Juniper Bean Co.Checking ••7718

Balance$212,480.16

To / FromCategoryAmount

SStripeSales+$41,200.00

KKettle & CrateCOGS−$12,400.00

GGustoPayroll−$2,140.00

Verifiers

Checks on the end state the agent leaves in the sims

6/7 pass

Same line items and layout as Q2pass

Revenue $3,180 high: CAD counted at face valuefail

Q3 P&L sent to the CFOpass

Final .xlsx attachedpass

Refunds netted out of revenuepass

All 47 Q3 transactions categorizedpass

Payroll kept in OpEx, not COGSpass

Simulators used

The environments run on 210 simulated apps, from Gmail and Excel to open-source ERPs, health records and helpdesks.

Every task contributor goes through a rigorous pipeline to ensure they meet our quality bar.

Works at a relevant organization · uses AI heavily in their job · 3+ years in their domain

A communication skills assessment and a domain expertise assessment

An introduction to RL task creation and how to create high-quality tasks, then Plato platform training

Domain experts and a Plato RL task specialist check accuracy, relevance, task quality and fit with RL objectives

The 300 tasks span 10 domains and 36 roles.

Finance & Accounting: 75 tasks (25%) FP&A Analyst — budgets & forecasts: 25 tasksStaff Accountant — close & recon: 19 tasksWealth Advisor — client portfolios: 12 tasksCredit Analyst — lending & covenants: 11 tasksTechnical Accountant — audit support: 8 tasks Engineering & Design: 64 tasks (21%) Quality Engineer — inspection & release: 22 tasksDesign Engineer — CAD & analysis: 10 tasksSystems Engineer — MEP & civil review: 10 tasksManufacturing Engineer — process & BOM: 10 tasksIndustrial Engineer — productivity: 7 tasksProject Architect — RFIs & submittals: 5 tasks Legal & Compliance: 42 tasks (14%) Corporate Counsel — contracts & deals: 11 tasksParalegal — litigation & settlements: 11 tasksCompliance Analyst — controls & AML: 8 tasksEstate Attorney — trusts & estates: 5 tasksFamily Law Attorney — custody memos: 4 tasksClaims Counsel — coverage memos: 3 tasks Healthcare Administration: 28 tasks (9%) Quality Manager — clinical governance: 10 tasksPractice Manager — scheduling & staffing: 8 tasksNurse Reviewer — medication safety: 6 tasksMedical Biller — eligibility & auths: 4 tasks Marketing: 23 tasks (8%) Marketing Analyst — performance data: 8 tasksDigital Marketer — paid, web & ads: 8 tasksMarketing Manager — campaigns & comms: 7 tasks Insurance / Claims: 23 tasks (8%) Account Manager — policy servicing: 12 tasksClaims Examiner — claims & bordereaux: 6 tasksUnderwriter — renewals & risk: 5 tasks Software & IT: 19 tasks (6%) Platform Engineer — incidents & code: 8 tasksSecurity Analyst — access & incidents: 6 tasksData Engineer — pipelines & quality: 5 tasks Sales & Revenue: 12 tasks (4%) Sales Ops — orders & fulfillment: 8 tasksBusiness Development — pipeline: 4 tasks Operations & Management: 7 tasks (2%) PMO Lead — planning & resourcing: 4 tasksProduct Ops Lead — launch gates: 3 tasks HR & People Ops: 7 tasks (2%) L&D Coordinator — training: 4 tasksRecruiter — hiring reviews: 3 tasks300tasksFinance & Accounting7525%

Engineering & Design6421%

Legal & Compliance4214%

Healthcare Administration289%

Marketing238%

Insurance / Claims238%

Software & IT196%

Sales & Revenue124%

Operations & Management72%

HR & People Ops72%

Finance & Accounting

FP&A Analyst · budgets & forecasts25

Staff Accountant · close & recon19

Wealth Advisor · client portfolios12

Credit Analyst · lending & covenants11

Technical Accountant · audit support8

Engineering & Design

Quality Engineer · inspection & release22

Design Engineer · CAD & analysis10

Systems Engineer · MEP & civil review10

Manufacturing Engineer · process & BOM10

Industrial Engineer · productivity7

Project Architect · RFIs & submittals5

Legal & Compliance

Corporate Counsel · contracts & deals11

Paralegal · litigation & settlements11

Compliance Analyst · controls & AML8

Estate Attorney · trusts & estates5

Family Law Attorney · custody memos4

Claims Counsel · coverage memos3

Healthcare Administration

Quality Manager · clinical governance10

Practice Manager · scheduling & staffing8

Nurse Reviewer · medication safety6

Medical Biller · eligibility & auths4

Marketing

Marketing Analyst · performance data8

Digital Marketer · paid, web & ads8

Marketing Manager · campaigns & comms7

Insurance / Claims

Account Manager · policy servicing12

Claims Examiner · claims & bordereaux6

Underwriter · renewals & risk5

Software & IT

Platform Engineer · incidents & code8

Security Analyst · access & incidents6

Data Engineer · pipelines & quality5

Sales & Revenue

Sales Ops · orders & fulfillment8

Business Development · pipeline4

Operations & Management

PMO Lead · planning & resourcing4

Product Ops Lead · launch gates3

HR & People Ops

L&D Coordinator · training4

Recruiter · hiring reviews3

How we ensure world quality

We check every task on four dimensions to make sure it reflects real work, poses a meaningful challenge and is worth getting right.

The environment and prompt match the author’s real job.

In the Juniper Bean exampleThe accountant’s own stack: Excel, Gmail, Stripe and Mercury. Last quarter’s P&L sits in an email thread, where it normally would.

Everything needed to finish the task is in the environment.

In the Juniper Bean exampleStripe shows every CAD payment next to its USD-converted amount, so the right revenue figure is findable.

The work takes reasoning and several steps, not a lookup.

In the Juniper Bean exampleRevenue spans USD and CAD. Summing the Stripe export at face value overstates it by $3,180.

In the Juniper Bean exampleThe P&L goes straight to Elena, the CFO. An inflated revenue line reaches leadership unnoticed.

Reviews: two experts from the same occupation agree on each task’s realism and fairness, and the Plato team reviews it for environment defects, reward hacking and prompt/grading mismatches.

Partner with us to get access to any of these environments.

We’re hiring! Check out open roles or email hi@plato.so.