Financial Document Collection & Annotation Specialist

About OpenTrain AI

OpenTrain AI is the hiring and contracting organization for this role. OpenTrain is the #1 platform for building careers in AI training and data labeling — a place to start and grow work that directly shapes how modern AI systems behave.

We hire contributors worldwide for contract and part-time work across document, audio, image, and generative-AI projects. This role is part of that mission: practical, remote work that feeds state-of-the-art models.

About AI Training Work

AI training (data labeling/annotation) is the human side of machine learning: people prepare examples and corrected outputs models learn from. These projects are often 100% remote and flexible, letting you choose hours that fit your schedule.

Many projects require no prior experience; specialized tasks pay more for subject-matter knowledge. This role benefits from finance/wealth-management familiarity and involves careful, structured data work.

The Role

You will create a ground-truth dataset by pairing anonymized financial documents (PDF/XLS/images) with corrected, structured JSON outputs following a provided schema. The deliverable is a set of document + JSON “correction” pairs used to benchmark model performance and detect regressions.

This is a contract, part-time opportunity with a fixed price of USD 300 for the project. Work is remote and open worldwide; expected time commitment is less than 20 hours per week. You will use OpenTrain's internal proprietary tooling for annotation and submission.

• Employment type: Contractor, Part-time.
• Data type: Financial documents (PDF, XLS, images).
• Labeling software: Internal proprietary tooling provided by OpenTrain AI.

What You'll Do

Follow a clear, stepwise process to produce high-quality document + JSON pairs. OpenTrain will provide support for sourcing documents so you can focus on accurate extraction and formatting.

• Gather ~10 varied financial documents (support provided) and anonymize any personal data as required.
• Extract required client financial information: assets, liabilities, account/holding details, balances, currencies, dates, and identifying attributes.
• Populate the provided JSON schema/template accurately for each document.
• Validate JSON formatting and completeness before submission.
• Flag and document ambiguities or missing information using tags and notes per instructions.

Requirements

This is an intermediate-level task that requires accuracy with financial documents and comfort working with structured JSON. You must be detail-oriented and able to follow a schema precisely.

• Intermediate experience level as stated; proven accuracy with data extraction or annotation work.
• Ability to read and interpret common financial documents (statements, account reports, holdings).
• Comfort working with JSON schemas and validating JSON formatting.
• Basic familiarity with spreadsheets, PDFs, and common document/image formats.
• Wealth-management experience is a strong plus for sourcing and correctly labeling documents.

Who Should Apply

Apply if you want flexible, remote contract work that contributes directly to model benchmarking and improvement. Ideal candidates are precise, finance-savvy, and able to deliver high-quality structured outputs.

• Contractors seeking part-time, remote work (worldwide).
• Annotators or data specialists with attention to detail and experience handling financial records.
• Professionals with retail wealth-management or related finance experience are especially encouraged to apply.

How It Works

To apply, create or use your OpenTrain account and submit your profile and interest for this project. If selected, you'll receive project instructions, the JSON schema/template, and access to OpenTrain's internal tooling. OpenTrain will provide support for document sourcing where noted.

Deliverables are the anonymized documents plus validated JSON files and any tags/notes for ambiguities. The project pays a fixed price of USD 300. Specific payment timing and milestone details will be provided in the project agreement.

• Project compensation: Fixed price USD 300 (single project payment).
• Expected commitment: Less than 20 hours per week during the project.
• Work and deliverables are submitted through OpenTrain's proprietary tooling.