What the work actually looks like
You receive adolescent case materials — presentation, history, contextual detail — and produce a written risk assessment: what is self-harm without suicidal intent, what is ideation, what constitutes intent or acute risk, and how confident you are in that distinction and why. You then carry the formulation forward into a response: escalation threshold, safety planning along Stanley-Brown lines, means-restriction specifics, and the comorbidity and safeguarding factors that change the answer. A significant share of the work is cross-referencing your judgment against established guidelines and against other clinicians' assessments of the same case, and articulating where you diverge.
The end user of your writing is not a peer. These outputs train systems intended to support non-specialists, so guidance that reads as competent in a multidisciplinary team meeting may still fail here — hedged language, unexplained clinical shorthand, and recommendations that assume access to specialist services all get flagged. Clarity about the boundary between what a non-specialist can safely do and what must be escalated is the core of the job.
What the screen looks for
micro1 runs an AI-led interview before any human review. It probes verifiable specifics — where you trained, your registration, the age range and acuity of your recent caseload — and then presses on clinical depth: how you administer and interpret the C-SSRS, how you handle a disclosure that shifts mid-assessment, what you do when the adolescent's account and the parent's account conflict. Expect follow-ups that ask you to justify rather than restate. Written English is assessed directly, since the deliverable is prose.
Logistics
- Fully remote, contractor, asynchronous task delivery with per-task or per-hour accounting
- Roles typically fill within 48 hours; first tasks expected 24–48 hours after onboarding
- Volume is project-driven and can fluctuate; most experts run this alongside clinical work
- Observed band is $45–95/hr, weighted by credentials, subspecialty depth and calibration performance — not guaranteed
