Fleet AI Review: Remote Expert Contract Jobs & Pay

Fleet AI builds training and evaluation environments for AI agents — realistic tasks where an agent's work can be checked against expert standards. To create and grade that material, it recruits credentialed domain experts on remote contracts, separately from its full-time HQ roles in SF and NYC.

You apply to a specific contributor role (its careers page has a 'Remote' tab for these), and the contract engagements are fully remote. Unlike a crowd platform, Fleet hires each expert directly for a defined engagement.

In your field, you develop realistic problems and reference materials for AI training and evaluation, and you review AI-generated work to identify errors and explain their consequences — checking validity, safety and feasibility. For an engineering role that means real engineering problems; for law, it means assessing legal reasoning, factual support and citations against the relevant authorities.

It's genuine expert work, not simple rating — Fleet wants your professional judgment applied to hard, edge-case scenarios that test how well AI agents reason and produce valid results.

Rates are high for this kind of work and vary by field. The mechanical-engineering contract starts at $60–$100/hour and can rise to $150/hour based on deliverable quality; the legal contract pays $150–$200/hour, with top performers earning more. Initial rates are agreed before the engagement begins, and any increases are agreed before they take effect.

That transparency — set rates, confirmed before work — is a good sign, and the ceilings are among the higher ones in AI-training work, reflecting the credentials required.

You apply to the remote role that matches your expertise. After an initial conversation, you may be invited to complete a take-home exercise — for the engineering role, one designed to take no more than about three hours — that assesses your technical approach and judgment.

Because these are credentialed roles, expect to demonstrate real professional expertise. The screening is more involved than a typical gig sign-up, which fits the pay level.

Fleet AI presents as a serious, well-resourced AI lab: a real HQ in SF with a NYC satellite, a full benefits package for staff (equity, health, 401k, gym membership), a dedicated support channel for its fellows, and direct, contract-based hiring. The contributor roles have clear scopes and stated rates. It doesn't ask contributors to pay anything.

The main caveat is that it's a newer company, so there isn't a long public track record of contributor payouts yet. As with any newer engagement, confirm your rate and terms in writing before starting, and keep records of your work.

Fleet is an excellent fit for credentialed professionals — experienced engineers, practising or former attorneys, and other domain experts — who want high-paying, flexible remote work that uses their actual expertise. The rates reward depth, and the work is intellectually serious.

It's not for generalists or entry-level applicants: the roles gate on real credentials and a take-home, and only a couple of fields are open at any time. If your field isn't currently listed, there may be nothing to apply to right now.

Yes, if you qualify. Fleet AI offers some of the better-paid remote AI-training work around ($60–$200/hour), from a legitimate, well-resourced lab, with transparent rates agreed up front. The caveats are modest: it's newer, selective, and runs only a few contributor fields at a time.

If you're an engineer, lawyer or other credentialed expert, it's well worth applying to an open Fleet Fellow role — just confirm your rate and engagement terms before you begin.