The work

An AI lab is training models to perform real enterprise sales work — the multi-step, judgment-heavy sequences that only someone who has carried a quota can evaluate credibly. Your job is to be the ground truth. That means working tasks end-to-end inside sandboxes that mirror a live sales stack (CRM, communication, billing tools), producing the expert trajectory the model learns from, then turning around and grading the model's attempts at the same tasks.

A typical week mixes three kinds of work:

  • Golden trajectories: completing prospecting sequences, pipeline hygiene passes, forecast reporting, and customer-facing documents at the quality a senior sales leader would sign off on.
  • Workflow audits: reviewing what the AI did across a multi-step task and articulating precisely where its judgment diverged from a strong seller's — not just that the output was weak, but which decision went wrong and why it matters commercially.
  • Rubric work: pressure-testing the scoring criteria so they capture what actually separates excellent selling from competent selling, and flagging where a rubric rewards surface polish over substance.

What the screen looks for

Mercor's AI-led interview probes for specifics, and follow-ups get concrete fast. Expect to be asked about deal sizes, cycle lengths, quota attainment by year, and the mechanics of how you ran your pipeline — vague summaries of "exceeding targets" don't survive the second question. The bar is roughly a decade of hands-on enterprise selling; ICs, front-line managers, and VPs are all in scope, but the common requirement is that you personally did the work recently enough to reconstruct it. Day-to-day fluency with Salesforce, Google Workspace, Microsoft Office, Jira, and Confluence is assumed, plus the comfort to pick up HubSpot, Zoho, Slack, Notion, Stripe, Intercom, Aircall, or Teams on short notice. Strong writing matters as much as sales instinct — much of the output is prose and slides someone else will be judged on. Experience with two-sided AI-powered B2B SaaS creator marketplaces and outcome-based campaign targets is called out as a differentiator worth mentioning.

Logistics

This is structured W-2 employment with Cincinnatus LLC acting as employer of record, with placement into a lab's extended workforce — not a freelance or task-marketplace engagement. Cincinnatus administers payroll, benefits, and onboarding; Mercor is the discovery channel. Fully remote within the United States, with part-time or full-time commitments depending on the pod, and enough overlap with the client team's hours to participate in calibration and rubric discussions. Rates in the $60–90/hr band reflect what has been observed on comparable postings and are not guaranteed.