Legal Expert — Tier 1 Specialist (Project Donna) | $150-$156/hr Remote
Overview
This contract role places you inside a blinded human-preference study run by a leading legal AI developer. You will receive a synthetic legal matter with its underlying documents, then evaluate six AI-generated deliverables labeled A through F. Your task is to rank those outputs through five head-to-head comparisons against a common comparator, and justify each choice in a short written rationale. The project focuses on the scarcest specialisms, so only lawyers with deep current practice in sanctions and export controls, structured finance and securitisation, or life sciences and pharmaceutical regulatory will qualify. The work is remote, spans roughly 24–30 hours, and demands strict confidentiality under NDA.
What You'll Do5
- 1Read through a synthetic legal matter and review all attached source documents to understand the context.
- 2Assess six blinded AI outputs (A–F) for accuracy, completeness, sound judgment, and practical usefulness.
- 3Conduct five direct comparisons between each output and a common comparator, choosing the stronger result.
- 4Write a concise, specific rationale for every preference, citing concrete reasons tied to legal reasoning.
- 5Adhere strictly to the assignment timeline and follow written instructions without external clarification.
Requirements5
- 1Hold a current legal qualification and possess deep, hands-on experience in one of the three listed specialisms.
- 2Demonstrate ability to evaluate legal work product holistically, considering correctness, thoroughness, and real-world applicability.
- 3Produce precise written English that clearly explains your comparison decisions in a few sentences.
- 4Work independently under a deadline, managing 24–30 hours of review time across the engagement without hand-holding.
- 5Sign an NDA and commit to not using any external AI tools on project materials.
Who Should Apply
This role suits a practicing lawyer with recent, substantive work in sanctions, structured finance, or life sciences regulation who can judge AI output with the same rigor they apply to a junior associate's memo. You need to be comfortable making quick, defensible calls and articulating them in crisp prose. If you lack hands-on experience in one of these narrow fields, or if you are more comfortable with general litigation or corporate work, this is not a good match. Candidates often miss out by submitting vague qualifications or failing to show how their current caseload aligns with the exact specialism, and by underestimating the need for concise written justification under time pressure.
Salary Insight
The rate is set at $150–$156 per hour, which sits above typical freelance legal review work, reflecting the rare subject-matter depth required. This pay level matches experienced lawyers who can hit the ground running without training, not junior or generalist practitioners.
Location
Required Skills
Application Tip
In your application, name the specific specialism you practice daily and give one concrete example of a matter you handled recently, such as a sanctions compliance review or a securitization deal. The evaluators are looking for evidence that you can judge nuance, not just list credentials, so highlight any experience comparing or critiquing legal work product.
See NearSkill jobs more often in your search
How your application is processed
1Application received
Your resume and details are logged the moment you apply.
2ATS + eligibility screening
We check your profile against the role’s skills, seniority, and requirements.
3Employer sees qualified profiles only
Only candidates who clear screening move forward.
Similar open positions
Explore active roles that match your skills and interests.
Mercor
VerifiedLegal Expert — Disputes & International (Project Donna)
Act as a senior legal reviewer for a study run by a leading AI legal company, designed to gauge how expert attorneys judge machine-produced work. You will read a fabricated case file and then score six anonymized AI outputs, each labeled A through F, on the basis of accuracy and usefulness. The substantive focus sits squarely in dispute resolution, with an emphasis on international arbitration, sanctions, and cross-border matters like award enforcement and emergency relief. For every comparison you must write a short explanation of your choice, which directly shapes the AI tool's future behavior. The entire commitment spans about 24 hours and operates under a strict NDA.
Mercor
VerifiedLegal Expert — Disputes & International, UK (Project Donna)
UK disputes lawyers can join a blind human-preference study for a legal AI company. You will read a synthetic matter involving cross-border disputes, then assess six AI outputs (items A–F) against each other. Five direct comparisons plus a common baseline will shape your feedback. The work sits squarely in international arbitration, sanctions and export controls, and emergency interim relief.
Mercor
VerifiedLegal Expert — Regulatory, Privacy & Healthcare (Project Donna)
This is a blinded human-preference study run by Mercor on behalf of a legal AI company. You will read a realistic legal matter, review six AI-generated work products (A–F), and make five head-to-head comparisons against a common comparator output. For each preference, you'll write a short, specific rationale explaining your reasoning. The practice areas cover Rule 206(4)-7 adviser compliance, HIPAA and state clinical data privacy, ESG disclosure, and False Claims Act CID responses with Anti-Kickback and Stark analysis. All work happens remotely over about 30 hours under a strict NDA, with no external AI tools allowed on the material.
Mercor
VerifiedLegal Expert — Corporate & Capital Markets (Project Donna)
Mercor is running a blinded human-preference study for a legal AI company, and you can serve as the expert eye. You will read a fabricated corporate matter, then review six anonymized AI outputs labeled A–F. For each matter, you make five direct comparisons against a shared baseline and back each choice with a written rationale. The practice areas cover acqui-hire mergers, seller disclosure schedules, Rule 144A/Reg S closing opinions, and Series B financing. The whole engagement asks for about 24 hours of your time, all under a strict NDA.
Mercor
VerifiedLegal Expert — Finance & Real Estate (Project Donna)
Mercor is hiring experienced finance and real estate lawyers for a paid, remote study with a legal AI company. You will read a realistic synthetic matter and evaluate six AI-generated work products labelled A through F. The legal areas include senior secured credit, project finance, RMBS, and tenant-favourable commercial leases. Your role is to review these drafts, make five comparisons against a common baseline, and provide a short written rationale for each preference. The project runs about 30 hours under a strict NDA.
Mercor
VerifiedLegal Expert — Restructuring (Project Donna)
Act as an expert reviewer in a blinded preference study for a legal AI company, applying your restructuring and insolvency know-how to evaluate AI-generated work products. You'll examine a synthetic legal matter and its supporting documents, then assess six anonymized deliverables (A-F) across five head-to-head comparisons, recording your choices with clear written justification. The practice areas at stake are specific: Chapter 11 first-day motions, Section 363 sales and bid evaluation, and cross-class plan treatment. Your judgment matters, so expect to weigh correctness, completeness, and practical usefulness in every response. Expect about 18 hours of work across the engagement, with one matter in this pod requiring more time than the others, and note that everything is under a strict NDA that bars using external AI tools on project material.