What the work involves
The project is about capturing operational judgment that isn't written down anywhere. Day to day you might be asked to review a CRM data model a model has proposed and explain where it breaks under multi-product bundling; to write a walk-through of how you configured lead routing or renewal stages and why you rejected the obvious alternative; or to arbitrate two conflicting policies — say, a finance rule on revenue recognition timing versus a CS rule on when an account counts as churned — and document the tradeoff in enough detail that a model could learn the reasoning, not just the verdict.
Tasks are scenario-based and written-heavy, with some verbal walk-throughs. Expect to produce artifacts like:
- Critiques of model-generated workflows, flagging compliance gaps and reporting side effects
- Annotated process designs showing decision points and what you deliberately left stable
- Written rationales for data integrity fixes, including what the bad data was doing to downstream reporting
What the screen looks for
micro1 runs an AI-led interview, and it presses on specificity. The filter is hands-on ownership versus oversight: the listing explicitly excludes management-only exposure. Screeners want the object names, the field-level decisions, the CPQ or billing tool you were in weekly, and the regulated context — fintech, healthtech, insurtech — where a process choice had a compliance consequence. Vague fluency in "RevOps best practices" reads as secondhand. No AI or annotation background is required; domain depth is the credential.
Logistics
Contractor, fully remote, asynchronous task queues with occasional scheduled calls with project stakeholders. Volume is typically part-time and variable rather than a fixed weekly commitment — most contributors fit it around existing work. Observed pay for this listing sits at $50–100/hr, varying with depth of experience and task type; rates are as reported, not guaranteed.
