What the work involves

This is domain-expert data work dressed in the vocabulary of your day job. On a typical task you might be asked to write a cold outreach sequence to a materials engineer at a specialty coatings manufacturer, draft the discovery questions you'd use to establish budget and authority on a $200K polymer supply deal, or produce a written response to a procurement objection about lead times and certification. Other tasks are evaluative: two model-generated sales emails or call scripts are placed side by side and you decide which one a real buyer would answer, then explain the reasoning in enough detail that a non-salesperson could apply your standard.

The rubric-heavy parts matter most. Reviewers care whether the model invents technical specs, misuses terminology like tensile modulus or ASTM designations, over-promises on compliance, or pitches features to a stakeholder who cares about throughput and total cost. Your job is to catch that and articulate it — vague scores without written justification get sent back.

What the platform screens for

  • Verifiable sales history in technical or scientific materials — chemicals, polymers, composites, lab consumables, industrial coatings, semiconductors, metals. Named accounts, deal sizes, and cycle lengths come up in follow-ups.
  • Depth under probing. The AI interviewer will push past your first answer on a qualification framework or objection to see whether you actually ran the motion or read about it.
  • Writing quality. Much of the deliverable is prose, so clarity and concision are assessed directly from your interview responses and any sample task.
  • Evaluation temperament — willingness to be specific about why an output is wrong, not just that it is.

Logistics

Fully remote, contractor engagement, asynchronous task queues with per-batch deadlines rather than fixed shifts. Most contributors report 10–20 hours a week with volume that fluctuates by project phase; some weeks the queue is thin. Observed pay for this listing is $45–85/hr, typically set by demonstrated depth and how narrow your materials specialization is — not guaranteed, and rates can be revisited per project. No AI or machine learning background is expected.