Skip to content

For anyone working with AI · Starting with tech

Graded on how you think and steer the work

Anyone can use AI.
We measure who delivers.

Take on real-world missions with an AI agent and get a full breakdown—from how you directed the AI to what you shipped.

Practise on your own, or share your result with hiring teams.

  1. 1Choose a missionRead the brief and terms first
  2. 2Work with an AI agentRead, decide, and verify
  3. 3Read each piece’s resultSee if it met the bar
Explore a sample result

Why the old signals say less

Good-looking work doesn't tell you who's good anymore

  • CV

    AI can polish it for anyone

  • Clear-cut quiz

    AI handles many clearly specified tasks

  • Take-home

    It can look good either way. You can't see who decided what

So we think the output alone isn't enough anymore.

What you really want to know

Does this person know what's right,
and will they catch the AI when it's wrong?

The result

See the result, plus how it was done

  1. The first line states the rule in one sentence: every piece reached its required level, 7 of 7, or which piece hasn't yet and the level it needs.

  2. A piece has levels drawn from C1 to C5; it need not have all five. Each piece has its own required level.

  3. Time used, out of the time given.

TimelineSample data
Work period+18 −4
  1. Message to AI

    Read the request handler and trace the retry path.

  2. Commandbun test requests.test.ts
  3. src/requests.ts

    Captured change available

    Diff

The timeline shows recorded work; incomplete coverage is marked separately.

Sample result: the verdict, each piece's levels, and attempt details.

The first line states the rule in one sentence: every piece reached its required level, 7 of 7, or which piece hasn't yet and the level it needs.

TimelineSample data
Work period+18 −4
  1. Message to AI

    Read the request handler and trace the retry path.

  2. Commandbun test requests.test.ts
  3. src/requests.ts

    Captured change available

    Diff

The timeline shows recorded work; incomplete coverage is marked separately.

Sample result: the verdict, each piece's levels, and attempt details.

What we measure

A finished file leaves questions

A submission shows what was delivered. A captured process gives a reader something to follow. This illustration compares records, not scores based on when AI was first used.

Made-up example allocation/ledger.ts

Visible tests pass for both; the diffs look alike

  1. Person 1 Checks recorded
  2. Person 2 No check captured
  • Read the evidence
  • Steer the AI
  • Check what worked
  • Submit
  1. What was submitted

    The accepted work and mission version identify what was evaluated.

  2. The captured process

    Messages, commands and file changes within the retained record.

  3. Questions to follow up

    A reader can follow available evidence with questions rather than inferring everything from one file.

Missing records remain gaps. They do not prove no work happened. Evaluation follows the task requirements.

Read a work-shaped brief

One symptom, multiple clues to trace

Sample mission topics—not the live catalog. Read the full brief and deliverables before choosing.

Duplicate payment

Duplicate requests and unintended side effects

  • 75 min
  • 4 domains
  • Sample: Defense required

Inventory reconciliation

A balance that doesn't match its records

  • 60 min
  • 3 domains
  • Sample: Defense required

Service recovery

An issue that takes multiple sources to trace

  • 90 min
  • 5 domains
  • Sample: Defense required

Illustrative times and piece counts only

Separate symptoms from requirements before editing

See a single error → Ask AI for an immediate fix

The requirement is still unverified

  1. Inspect terms and supplied data

  2. Check the repair against the deliverable

There is reasoning and a check to inspect

From brief to result

Understand the work. Then work with AI.

Check the brief and rules, do the work in the workspace, and submit for grading.

  1. 01

    The brief defines the deliverable

    Read the published brief and mission version. Some internal runtime details aren't shown upfront.

    Illustrative brief materials
    • brief Situation and deliverables
    • workspace Starting work files
    • terms Time, credits and acknowledgements
  2. 02

    Rules are locked to your attempt

    Each version and preset locks the timer, AI budget, allowed models, tool policy, starter files, and grader. Only the model you pick and its outputs vary.

    Read the work in the brief
    • Situation The work to handle
    • Deliverables The required output
    • Content language As published for the mission
  3. 03

    Submit, then read piece by piece

    Check whether each piece met its required level. Doing extra on one piece won't make up for missing another.

What a result refers to

One attempt, one set of terms

Read the version, submission and piece levels together to understand the result’s scope.

  1. Choose the version

    Read that version’s brief to see what this attempt must deliver.

  2. Read the terms

    Read the acknowledgements to understand this attempt’s rules.

  3. Work in the workspace

    Read files, direct AI and inspect changes within the mission’s rules.

  4. Submit the work

    Review the intended work and follow the mission guide for acceptance.

  5. Read the result

    Compare each piece’s achieved level with its required level.

The pass rule

Every piece

Each piece has its own bar

A piece that meets its bar cannot compensate for one that does not.

Read the example

C1 is the core. Deeper levels do not count until it passes.

The complete explanation and worked example are on How we measure.

  • Version and preset

    Terms belong to the started attempt, not a later choice based on its outcome.

  • Missions have versions

    A name can span several versions. Read the version beside the result.

  • Per-piece requirements

    Every piece has its own required level.

  • State and history

    A hold is not a fail; corrections keep their history.

Read the full mechanism

Read each source for what it shows

Give finished work more context

Different sources answer different questions. This compares information, not hiring outcomes.

Written background Final files Conversation Mission result
What to read Described experience The submitted artifact The explanation Each piece’s level
Further context Check other sources Compare with requirements Read beside the work Captured, retained evidence
The reviewer’s role Assess relevance Inspect quality Interpret the explanation Read terms and limitations

A result does not guarantee employment or replace human judgment.

  • Separate requirements

    See which pieces reached their required level.

  • Read the measurements

    Pieces define what the result covers.

  • Inspect recorded process

    Open a timeline and changes within access and retention.

  • Explicit data boundaries

    Unmeasured and low-scoring are different.

  • Reviewable results

    Corrections keep a visible history.

  • People decide

    Read alongside other job-relevant evidence.

Put results to work

Different teams ask different questions

  • CandidatesDiagram
    Question
    Read the work before talking
    Context
    Work relevant to the purpose
    Continue
    The organisation guide has the details

    Candidates Read the work before talking

    Review shared results alongside the rest of your hiring process.

  • TeamsDiagram
    Question
    Plan practice from available data
    Context
    Work relevant to the purpose
    Continue
    The organisation guide has the details

    Teams Plan practice from available data

    In team rounds, check result coverage before reading the charts.

  • ClassesDiagram
    Question
    Put feedback beside the work
    Context
    Work relevant to the purpose
    Continue
    The organisation guide has the details

    Classes Put feedback beside the work

    In classes, teachers can review shared work and leave feedback without altering the graded result.

Fits the hiring you have

Place work evidence in your process

Choose a position to explore an illustrative flow. Add mission evidence while your team reads its limits and makes the decision.

Before the résumé

Job post → OperatorRank → Résumé review → Interview / live coding → Decision

Read an accessible result alongside the application, without automatic rejection.

Instead of the take-home

Job post → Résumé review → OperatorRank → Interview / live coding → Decision

For an assessment step, match the brief and terms to the work.

Before the interview

Job post → Résumé review → Take-home → OperatorRank → Interview / live coding → Decision

Read permitted evidence before talking to understand the attempt in context.

Finish with us

Job post → OperatorRank → Decision

Use it as one source alongside other evidence, with people deciding.

Practice on your terms

Revisit your past work

Each result keeps a record of how you worked, so you can see what to practise next.

  • Read each piece

    Compare the required and achieved level.

  • Inspect your own process

    Review saved prompts, commands, and file edits.

  • Keep privacy boundaries clear

    Practice reaches a company only when you share it

  • Get free credits

    Verify with Google, then check your balance and expiry under Credits.

Know the boundaries

Clear rules, and clear limits on who sees your work

Records within the work

  • Captured conversation and commands
  • Retained files and changes
  • Results and review history

The workspace does not use

  • Camera
  • Microphone
  • Screen capture
  • Read the issued state

    A held result has no verdict to interpret.

  • Use the allowed model menu

    The menu and budget belong to the preset.

  • Read unmeasured data correctly

    Unmeasured doesn't mean a low score.

  • Practice is not public

    Organisations only see what your permissions and consent allow.

  • A gap stays a gap

    Missing evidence does not prove no work happened.

  • Results are provisional and correctable

    Check the review deadline on the result page; any correction keeps its history.

Who it's for

For people who do the work, and people who hire them

For companies

See real work before you talk

Read each result piece by piece. No camera proctoring.

TimelineSample data
Work period+18 −4
  1. Message to AI

    Read the request handler and trace the retry path.

  2. Commandbun test requests.test.ts
  3. src/requests.ts

    Captured change available

    Diff

The timeline shows recorded work; incomplete coverage is marked separately.

For you

Let your work speak, not only your CV

Practice results are not public. Our team and providers process data as the Privacy notice explains.

Questions

Still wondering about something?

Can't find what you're after? Write to us at support@operatorrank.com

Does it cost anything?

Missions use credits; Pricing explains the cost.

See pricing and credits →

Is this only for tech?

We're starting with software tasks; browse the catalog for live missions and their content language. The same approach fits other real work with AI: spot the issue, check the evidence, make the call, and verify what the AI did.

See frequently asked questions →

Can't the AI just do it all?

Use the AI within the mission's budget and rules. But you pass only when every piece reaches its level: many fixes aren't enough if a piece still misses its requirement.

See how to read a result →

How fast is the result?

The result page gives an estimate when it can. Large missions take longer to check, and queues can add time. You can close the page; you'll get a notification in the app.

See how a mission runs →

Does the AI decide who gets hired?

No. We provide the work evidence; people on the hiring team read it and decide.

See the organisation workflow →

You are on-call. Good luck.

Choose work to try and read its brief before starting.