What the work involves

You'll be given model outputs on tasks a RevOps, SalesOps, or CS Ops practitioner would recognise: writing a lead routing spec, diagnosing why bookings and billings disagree, building a stage-conversion forecast, designing a quota model for a mixed new-logo/expansion team, or drafting a health-score definition that isn't just a login-count proxy. Your job is to judge whether the output is correct, complete, and defensible — then rewrite it to the standard you'd expect from a senior operator, with reasoning attached. Some batches instead ask you to author prompts and reference answers, so the model has ground truth to be graded against.

The grading is unforgiving about specifics. "This forecast looks reasonable" is not a review; "the model applied a blended win rate across segments where enterprise and SMB conversion differ by 3x, so the commit number is inflated" is. Reviewers who can name the failure mode and the fix consistently get more work than reviewers who only mark a score.

What the platform screens for

  • Systems you've actually administered. Expect follow-ups on object models, field-level dependencies, reporting limitations — the kind of detail that separates someone who's built in Salesforce from someone who's read about it.
  • Metric definitions under pressure. Net revenue retention, pipeline coverage, ramped rep capacity, ARR vs. bookings vs. recognised revenue. The screen will ask you to define one and then push on the edge case.
  • Judgment when the answer is ambiguous. Many RevOps questions have several defensible answers; the screen wants to see you name the tradeoff rather than assert a single truth.
  • Written clarity. Your rationale is the deliverable. Long, hedged prose scores worse than short, specific prose.

Logistics

Fully remote and largely asynchronous. Work arrives in batches with turnaround windows rather than fixed shifts, and volume fluctuates with client demand — treat it as a supplement to a primary role rather than a stable full-time load. Observed rates for this category run $40–80/hr, positioned by depth of operating experience and the tooling you can speak to credibly; rates are as observed and not guaranteed. The screening interview is conducted by an AI interviewer, recorded, and typically takes 20–40 minutes.