Document Extraction Specialist — Legal & Finance

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect people to cutting-edge annotation work that helps shape how modern AI systems behave.

OpenTrain AI is hiring contractors for this role. You will join a growing community of contributors who perform the human-focused tasks that make AI useful and reliable.

About AI Training Work

AI training (data labeling/annotation) is the human side of building intelligent systems: people read, tag, correct, and structure the examples models learn from. This project focuses on document-level extraction and accuracy-heavy validation.

These projects are often flexible and remote, and they let you contribute specialist knowledge (legal, finance, accounting, compliance) to improve model outputs while working part-time and on a schedule you choose.

The Role

You will extract structured information from PDF and DOCX documents (legal contracts, patents, financial records, compliance documents, and similar). Each task presents three AI-generated draft outputs; you will select the best draft, validate and correct every field, and produce a final JSON that conforms exactly to the provided schema.

This is a part-time contractor role (less than 20 hours per week). Pay is hourly, USD 8.00–11.20 per hour. OpenTrain AI hires and contracts contributors worldwide for this work.

• Work type: Document extraction and annotation (DOCUMENT data type).
• Labeling focus: Validate/edit AI drafts to a JSON schema; capture repeated entries and nested structures.
• Employment: Contractor, Part-time; worldwide contributors welcome.

What You'll Do

• Read source PDF/DOCX documents thoroughly before extracting data.
• Review three AI-generated draft outputs, select the best one, and edit it rather than starting from scratch.
• Follow the provided JSON schema precisely (required/optional fields, types, arrays, nested objects).
• Validate every field against the source document and correct errors or omissions with high accuracy.
• Capture repeated or dynamic entries such as line items, multiple clauses, and lists.
• Set truly missing required fields to null per schema instructions.
• Write concise, original summaries in your own words where the schema requires them.

Requirements

• Background in at least one of: law, finance, accounting, compliance, or data analytics.
• Prior experience with RLHF/annotation, data analysis, or accuracy-heavy data entry.
• Solid JSON schema literacy and comfort with nested structures and data types.
• Strong attention to detail when working with long, dense documents.
• Near-native English comprehension (C1/C2).
• Experience level: Intermediate.

Who Should Apply

Apply if you enjoy detailed, precision-focused work and have domain familiarity with legal, financial, or compliance documents. Ideal contributors are careful readers who can translate dense text into clean, validated structured data.

This role fits editors, paralegals, accountants, compliance analysts, data analysts, or annotators with schema experience who want flexible, remote, part-time work.

How It Works & Payment

Each annotation task is completed in an annotation panel with three AI draft outputs as starting points. Your deliverable is a validated JSON object that follows the schema exactly. When required fields are truly absent, set them to null as instructed.

Pay type: Pay-per-hour. Hourly rate range: USD 8.00 (minimum) to USD 11.20 (maximum), with typical assignments under 20 hours per week. You will be engaged as a contractor by OpenTrain AI.

• Data types: DOCUMENT. Label types listed for this project include RLHF, data collection, and programming/coding-related annotation.
• Worldwide applicants accepted; apply through your OpenTrain contributor account.