micro1 is contracting corporate attorneys to build and grade training data for AI systems working on M&A, private equity, capital markets, and commercial litigation problems. You write realistic legal scenarios, annotate transactional and litigation documents, and critique model outputs where the legal reasoning breaks down.

Summary, eligibility guidance and screening preparation are prepared by Sidequest from the platform's posting. Check the source listing before applying.

What the work actually involves

This is not advisory work for a client — it is data work for a model. On a typical task you might be handed a fact pattern involving a stock purchase agreement, asked to write the analysis a competent corporate associate would produce, then asked to grade two model attempts against it. Other batches lean toward document annotation: marking up indemnification provisions, MAC clauses, earnout mechanics, or disclosure schedules so the model learns what each moving part does and where the negotiation leverage sits. Litigation batches ask you to distill pleadings and briefs into structured representations of facts, claims, and arguments.

The recurring judgment call is whether a model answer is wrong or merely differently drafted. A lot of legal work has multiple defensible formulations, and reviewers who flatten that into personal drafting preference produce noisy data. Conversely, answers that read fluently but misstate a fiduciary standard, cite a Delaware rule that does not exist, or quietly swap buyer-favorable for seller-favorable language need to be caught and explained. Your written rationale is as much the deliverable as the score — it is what the training pipeline consumes.

What the screen looks for

micro1 runs an AI-led interview. Expect it to verify the J.D. and bar admission, then push on transaction and matter specifics: what you drafted, which side you represented, deal size and structure, what got negotiated. Vague seniority claims get probed until they either resolve into detail or collapse. Candidates with a technical background, a CS degree, or prior annotation experience are explicitly preferred, but domain depth is the gate — no AI experience is required.

Logistics

Fully remote contractor engagement, asynchronous, no fixed shift
Observed band is $140–350/hr, varying with specialty, seniority, and task type; not guaranteed
Volume is project-driven and comes in batches — sustained availability matters more than any single week
Work is written and independent, with periodic calibration against technical specialists and other reviewers

Corporate Attorney

What the work actually involves

What the screen looks for

Logistics

Interview prep — 10 questions