Science AI Trainer, $75-$100/hour

LinkedIn; United States; Remote; USD 83,000 - 165,000; Part time

Job summary

LinkedIn is seeking skilled Computational Science Experts to join as project consultants in its AI Labor Marketplace. Work is part-time, contract-based, remote for U.S.-based experts. You will design terminal-based scientific tasks, build Docker environments, and create verification tests to evaluate advanced AI systems. Candidates have 5+ years hands-on experience or a PhD/MS and strong programming ability. Docker and GitHub experience required.

Qualifications

• 5+ years of hands-on experience or advanced research specialization, such as a relevant PhD or MS with substantial computational work.
• Deep expertise in at least one scientific field (chemistry, biology/genomics, physics, mathematics, robotics, linguistics, or Earth sciences).
• Strong hands-on programming ability and ability to independently implement problems they design.
• Experience using code to perform substantive scientific or mathematical work (computational modeling, simulation, genomics, or research software development).
• Strong debugging and automated testing ability.
• Experience with Git and GitHub workflows.
• Experience building or working with Docker-based environments.
• Familiarity with automated build and test workflows.
• Ability to turn domain-specific scientific requirements into objective, machine-verifiable success criteria.

Responsibilities

• Design technically difficult but fair computational problems grounded in their scientific specialty.
• Translate scientific requirements into precise, self-contained task specifications.
• Build reproducible, containerized task environments using Docker.
• Implement a complete reference solution using appropriate scientific programming languages and libraries.
• Create hidden automated tests or verifiers that evaluate scientific and computational correctness.
• Account for edge cases, numerical behavior, constraints, and plausible but incorrect approaches.
• Reason adversarially about how an AI agent might produce an output that appears valid without solving the underlying problem.
• Validate that the reference solution passes evaluation while the baseline environment does not.

Skills

Programming

Scientific reasoning

Automated testing

Problem design

Education

PhD or MS with substantial computational work

Tools

Docker

Git

Job description

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 skilled Computational Science Expert to work as a project consultant 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:

Contributors will create challenging, terminal‑based scientific tasks used to evaluate advanced AI systems. We are seeking scientists with deep expertise in areas such as chemistry, biology and genomics, physics, mathematics, robotics, linguistics, or Earth sciences who are also highly capable programmers.

Each contributor will translate their scientific expertise into a difficult but objectively verifiable computational problem. Contributors will define the problem, construct a reproducible Docker environment, implement a correct reference solution, and create hidden tests or verifiers that determine whether the problem was actually solved correctly.

This work is best suited to computational scientists, research engineers, and technically oriented researchers who regularly write substantial code. Scientific expertise without strong hands‑on programming ability is not sufficient for this project.

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:

• Design technically difficult but fair computational problems grounded in their scientific specialty.
• Translate scientific requirements and constraints into precise, self‑contained task specifications.
• Build reproducible, containerized task environments using Docker.
• Implement a complete and correct reference solution using appropriate scientific programming languages, libraries, or computational tools.
• Create hidden automated tests or verifiers that evaluate scientific and computational correctness.
• Account for relevant edge cases, numerical behavior, constraints, and plausible but incorrect approaches.
• Reason adversarially about how an AI agent might produce an output that appears valid without correctly solving the underlying scientific problem.
• Validate that the reference solution passes the evaluation while the untouched task environment does not.

Expected Outcomes:

• Complete terminal‑based computational science evaluation tasks.
• Clear and reproducible task specifications and environments.
• Working reference implementations.
• Dockerized environments with required scientific dependencies and configuration.
• Automated hidden tests or robust correctness verifiers.
• Validation that each reference implementation passes and the baseline environment does not.

Qualifications:

• 5+ years of applicable hands‑on experience or advanced research specialization, such as a relevant PhD or MS with substantial computational work.
• Deep expertise in at least one relevant scientific field, including chemistry, biology/genomics, physics, mathematics, robotics, linguistics, or Earth sciences.
• Strong hands‑on programming ability. Candidates must be capable of independently implementing the problems they design.
• Experience using code to perform substantive scientific or mathematical work, such as computational modeling, simulation, scientific computing, quantitative analysis, formal methods, genomics/bioinformatics, or research software development.
• Strong debugging and automated testing ability.
• Ability to distinguish numerical or scientific correctness from superficially plausible results.
• Experience with Git and GitHub workflows.
• Experience building or working with Docker‑based environments.
• Familiarity with automated build and test workflows.
• Ability to turn domain‑specific scientific requirements into objective, machine‑verifiable success criteria.
• Strong reasoning about edge cases, failure modes, and alternative implementations.

LinkedIn Data Security and Privacy Notice:

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