Patent Agent Expert

USPTO registered patent agents to build the claim drafting and prosecution scenarios frontier AI labs use to test model judgment. AfterQuery builds the expert evaluation data that frontier AI labs use to measure and improve their models. We're hiring USPTO registered patent agents as contractors to author and grade the patent prosecution scenarios those evaluations run on. What you'll actually do Author realistic patent scenarios, from prior art analysis through claim drafting and office action responses Write reference answers with the legal and technical reasoning behind each decision Grade model output for claim quality, procedural soundness, and technical accuracy Flag drafting that reads plausible but would fail at the USPTO Who this is for Patent agents registered to practice before the USPTO Practitioners with at least 2 years of patent prosecution experience Agents with depth in claim drafting, office action responses, and prior art analysis Clear writers who can document the reasoning behind a prosecution decision Contract and pay $90 to $125/hr , paid weekly via Stripe. Your rate is set from your experience, not negotiated down after you start. Remote, fully async. No fixed hours. Minimum 10 hours per week, no maximum. Scale up or down week to week. No screen recording, no keystroke monitoring, no idle timers. Independent contractor. Ongoing project work rather than single task batches. How our hiring process works Apply at /apply/patent-agent-expert. About five minutes. Resume plus a few structured questions. Resume review. Every application gets read against a published rubric and gets a decision. Written assessment. You may be asked to take a written assessment for further placement. Onboarding. Sign the contractor agreement, connect Stripe, and get platform access. Project match. You're matched to live work in your specialty, usually within a week of onboarding. About AfterQuery AfterQuery is an expert data lab. We build the evaluation and training data that frontier AI labs use to measure what their models can and can't do in real professional work. Our contributors are practicing professionals, not annotators, and the work is credited, reviewed, and paid at professional rates.