Project Overview:

Join a growing community of professionals advancing the next wave of AI. As an AI Trainer, you’ll play a hands-on role by analyzing and providing feedback on data to improve LLM performance, helping ensure that the next generation of AI technology is accurate and trustworthy.

We are seeking a group of skilled Domain Experts to work as a project consultants in our AI Labor Marketplace. This is not a full-time employment position — you will be engaged as an expert project consultant on a contract basis.

Location: U.S.-based experts only

Engagement: Part-time, project-based expert evaluation work

Work Type: Remote

Project Summary:

We are seeking experienced professionals in insurance/claims, finance/fintech, logistics, supply chain, compliance, and marketing operations to create challenging, realistic tasks based on problems encountered in their fields.

Domain expertise is the primary qualification. Ideal consultants have hands-on experience with real operational workflows, systems, regulations, processes, or data and understand the edge cases and constraints that make problems genuinely difficult in practice.

Consultants will translate that expertise into complete Terminal-Bench 4.0 evaluation tasks. Programming/coding expertise is helpful but secondary.

Consultant Engagement Terms:

This is a project-based consultant role. Consultants will be paid on a per-project basis; hourly rates are estimates based on anticipated completion time. Consultants control their own schedule, provide their own tools, and may simultaneously provide services to other vendors or employers (subject to those vendors’ allowances).

Responsibilities:

Contributors will:

• Identify challenging, authentic problems based on their professional domain expertise.
• Translate real-world workflows, business rules, constraints, and edge cases into clearly defined evaluation tasks.
• Develop precise task instructions that establish unambiguous requirements and expected outcomes.
• Build a reproducible Docker environment for each task.
• Develop a working reference solution, using coding assistance as needed.
• Create hidden tests that evaluate genuine correctness and are resistant to superficial shortcuts.
• Validate that the reference solution passes the evaluation while the untouched starting task fails.
• Review the completed task for domain accuracy, realism, and technical reproducibility.

Expected Outcomes:

• A complete, reproducible Terminal-Bench 4.0 task based on a realistic domain problem.
• Clear task instructions and evaluation criteria.
• A working Docker environment and reference solution.
• Hidden tests that reliably distinguish correct from incorrect solutions.
• Verification of expected pass/fail behavior.

Qualifications:

Significant hands-on professional expertise in at least one of the following areas:

• Insurance or claims operations
• Finance or fintech operations
• Logistics or transportation
• Supply chain operations
• Regulatory compliance, risk, or controls
• Marketing operations
• Direct experience working with real operational workflows, systems, regulations, processes, or data within your field.
• Ability to recognize domain-specific edge cases, common failure modes, and what constitutes a correct real-world outcome.
• Sufficient technical fluency to work in terminal-based environments and understand inputs, outputs, data, tests, and code.
• Comfortable using AI-assisted coding tools to help implement and troubleshoot solutions.
• Ability to independently validate whether a technical solution accurately represents domain requirements.
• Advanced coding or competitive programming experience is not required, but knowledge of coding/programming tools is highly preferred.

LinkedIn Data Security and Privacy Notice:

https://legal.linkedin.com/content/dam/legal/Global-Data-Privacy-Notice_LI_TalentCommunity.pdf