What the work actually involves

This listing is an intake, not a project. You complete a written questionnaire — roughly fifteen minutes if you write tightly, longer if you do it well — describing how your compliance practice actually runs. Expect prompts on how your week splits across alert review, escalation, documentation, and committee or regulator-facing work; which systems you live in (Actimize, Verafin, Fenergo, Quantexa, Mantas, Smarsh, Relativity, internal case managers); and where the decision points sit that a project designer reading a public AML overview would never guess at. You will also be asked to slice your own field the way insiders do — the difference between a correspondent-banking KYC seat and a wealth-onboarding seat, between L1 alert triage and an investigations analyst writing a SAR narrative, between testing controls and owning them.

The most valuable section is where AI already gets your work wrong. Concrete casework beats abstraction: a model that treats a name-screening hit on a common transliterated surname as a match, a summarizer that drops the predicate offense from a narrative, an entity-resolution output that collapses two distinct beneficial owners, a tool that cites the wrong reporting threshold or the wrong regulator for the product. Specific failure modes with the facts around them are what shape evaluation rubrics later.

What the platform screens for

Mercor's screen is looking for verifiable seat experience and the ability to describe mechanics rather than principles. Strong respondents name systems, thresholds, typologies, and escalation paths; weak ones restate the three lines of defense. Credentials — CAMS, CGSS, CFCS, CFE, CRCM, an ICA diploma, FINRA registrations — help with routing but deep hands-on experience without one is explicitly welcomed. Expect follow-up questions that push on any claim you make: if you say you handled sanctions escalations, be ready to explain who signed off, what documentation went in the file, and how a false positive was closed out.

Logistics

  • Fully remote and asynchronous; the intake is written, with no live interview required to complete it.
  • No fixed hours. Intake completion is unpaid or nominally compensated depending on the cohort — paid work comes from the engagements it seeds.
  • Observed compliance rates on the platform run $90–110/hr and scale with seniority; these are reported bands, not guaranteed offers.
  • Do not include client names, customer data, alert IDs, or anything else your confidentiality obligations cover. Sanitize examples before you write them.