Attorneys Needed for LLM Training (English Fluency Required)

About OpenTrain

OpenTrain is a central job board that aggregates AI-training and data-labeling work from many companies and platforms. Creating an OpenTrain account is free and applying to projects takes only a few minutes.

We help connect specialists and remote contributors with short- and long-term labeling projects so you can find flexible work that fits your schedule and expertise.

About AI Training Work

AI training (also called data labeling or human feedback) is the human process that helps AI systems learn to answer correctly, safely, and usefully. Tasks range from rating model outputs to preparing datasets used for fine-tuning.

This project focuses on legal content: you will interact with an internal large language model and label its answers so the model learns better legal reasoning and drafting.

The Role

You will work as a contract, part-time labeler who asks the client's internal LLM legal questions and then A/B labels the model's text responses. Labels are used both for evaluation and for fine-tuning.

Work is fully remote and available worldwide, with flexible scheduling under a less-than-20-hours-per-week commitment.

• Employment type: Contractor, Part-time
• Time commitment: Less than 20 hours/week (flexible)
• Data type: Text; Labeling software: client’s proprietary tool (OTHER)
• Label types: Evaluation/Rating and Fine-tuning (A/B labeling)

What You'll Do

You will pose or review legal prompts, read paired model responses, and assign preference/quality labels according to provided guidelines.

Tasks emphasize careful analysis of legal reasoning, accuracy, tone, and applicability to the question or document at hand.

• Ask or review legal questions and prompts fed to the model.
• Compare two model responses (A vs B) and choose the better answer based on rubric.
• Provide quality ratings or short feedback where requested for fine-tuning.
• Follow detailed annotation guidelines to ensure consistent labels.

Requirements

All requirements below are mandatory and will be verified during onboarding. Do not apply if you cannot meet them.

This is an entry-level labeling role in terms of previous annotation experience, but it requires formal legal education.

• Must hold a legal degree from an accredited university in an English-speaking nation (example: J.D. from NYU).
• Fluency in English required for reading, writing, and legal analysis.
• Comfort reviewing legal questions and documents across common practice areas.
• Reliable internet connection and ability to use web-based annotation tools.

Compensation & Schedule

Payment is per label at a fixed rate. There is no guaranteed weekly minimum and no weekly cap on work.

Current labelers on this project average about $2,000 USD per week, though individual earnings vary with time spent and throughput.

• Pay type: Pay-per-label
• Rate: $2.125 USD per label
• No weekly cap; earnings scale with labels completed
• Typical time: under 20 hours/week, schedule set by you

How It Works

Apply through OpenTrain and complete any onboarding checks the client requires. You'll receive project guidelines and access to the client's annotation interface.

Work by reviewing prompts and paired responses, then submit A/B preference labels and ratings. Feedback and occasional calibration tasks will help keep labeling consistent.

• Create an OpenTrain account and apply to this listing.
• Complete identity/credential verification and any short training modules.
• Do labeling tasks via the client’s web tool and submit labels for payment.
• Payments processed per client/contractor terms supplied on onboarding.

Who Should Apply

Experienced or newly graduated attorneys who want flexible, remote contract work and who enjoy close reading and comparative evaluation of legal writing should apply.

This role suits people who want to contribute to improving legal AI and prefer short, focused annotation tasks rather than continuous legal practice.

• Entry-level labeling experience is acceptable; legal degree is required.
• Good fit for attorneys seeking part-time, remote supplementary work.
• Not suitable if you lack a law degree from an accredited English-speaking institution.