AI Trainer (Software Engineering)

OpenTrain AI, Inc.; United States; Teletrabalho; USD 83 000 - 165 000; Tempo parcial

Resumo da oferta

OpenTrain AI, Inc. is seeking a remote expert software engineer to create realistic AI evaluation tasks, write precise instructions, and establish clear evaluation standards.

You will also build reproducible environments, reference solutions, and tests, and review AI-generated code for quality and robustness. This fully remote, part-time contractor role pays $60–$120 per hour equivalent, paid per accepted task that meets project specifications, with less than 20 hours per week.

Qualificações

• Expert-level software engineering ability with prior paid experience training, evaluating, benchmarking, or improving AI systems.
• Mid-Senior level experience expected.
• Experience building, evaluating, and benchmarking AI-related software tasks.

Responsabilidades

• Create demanding software engineering tasks for AI model evaluation.
• Write task instructions and define objective standards for correct solutions.
• Build reproducible environments, reference solutions, tests, and evaluation criteria.
• Review generated code for correctness, robustness, and reproducibility.
• Identify problems in implementation, debugging, dependencies, and performance.
• Contribute technical judgment to coding evaluations and benchmark-style tasks.

Conhecimentos

Software design

Debugging

Testing

Version control

Dependency management

Ferramentas

Docker

CI/CD

Command-line tools

Sandboxed environments

Linux

Descrição da oferta de emprego

Create software engineering tasks, build tests and reference solutions, and evaluate AI-generated code. This remote expert role pays $60 to $120 per hour equivalent, paid per task that meets the project specifications.

$60–$120/hr

equivalent, paid per accepted task

Compensation

Contract, part-time

Engagement

Remote

Location

• Algeria


• Argentina


• Austria


• Bahrain


• Bangladesh


• Belgium


• Brazil


• Canada


• Chile


• Colombia


• Denmark


• Egypt


• France


• Germany


• Ghana


• Greece


• India


• Indonesia


• Italy


• Japan


• Jordan


• Kazakhstan


• Kenya


• Kuwait


• Kyrgyzstan


• Lebanon


• Malaysia


• Mexico


• Morocco


• Netherlands


• Nigeria


• Oman


• Pakistan


• Palestine


• Peru


• Philippines


• Portugal


• Qatar


• Saudi Arabia


• Singapore


• South Africa


• Spain


• Sri Lanka


• Switzerland


• Taiwan


• Thailand


• Tunisia


• United Arab Emirates


• United Kingdom


• United States


• Uzbekistan


• Vietnam

The work

You will help train and evaluate advanced AI systems by creating realistic software engineering problems and reviewing the code they produce. Your work will turn practical engineering judgment into clear tasks, tests, reference solutions, and evaluation standards.

• Create demanding software engineering tasks for AI model evaluation.


• Write task instructions and define objective standards for correct solutions.


• Build reproducible environments, reference solutions, tests, and evaluation criteria.


• Review generated code and technical deliverables for correctness, robustness, completeness, and reproducibility.


• Find problems with implementation, debugging, dependencies, performance, testing, and environment setup.


• Contribute technical judgment to coding evaluations and benchmark-style tasks.

What it pays and takes

This is a fully remote, part-time contractor role focused on accepted task deliverables. Task availability and completion time may vary.

• Pay: $60 to $120 per hour equivalent, paid per task that meets the project specifications.


• Time: Less than 20 hours per week.


• Location: Remote, open to candidates in the countries specified for this role.


• Experience: Expert-level software engineering ability and prior paid experience training, evaluating, benchmarking, or improving AI systems.


• Required knowledge: Software design, debugging, testing, version control, and dependency management.


• Helpful evidence: Substantial repositories, open-source contributions, developer tools, systems, publications, benchmarks, or engineering write-ups.


• Helpful experience: AI training tasks, reference solutions, tests, graders, benchmark problems, technical submission reviews, long-horizon or agentic coding evaluations, Linux, Docker, CI/CD, command-line tools, sandboxed environments, or programming competitions.

About AI training work

AI training is the human work behind systems that learn from examples, including code reviews, model evaluations, and other labeled data. Experienced software engineers are paid for applying careful technical judgment to help make AI outputs more accurate, reliable, and useful.

Requirements

• Experience: Mid-Senior level

Build, debug, and evaluate full-stack software and AI-generated code across front-end and back-end systems. Earn $50 to $100 per hour equivalent, paid per task that meets the project specifications, for under 20 hours per week.

Build and improve open source software by adding features, fixing complex bugs, refactoring legacy code, and reviewing code. Pay is $50 to $100 per hour equivalent, paid per task that meets the project specifications.

Evaluate Rust repositories, tests, and bug-fixing tasks to measure how AI coding systems perform. This remote LATAM contract pays $25 per approved task and requires 20+ hours per week.

Evaluate C# open-source repositories and test how well AI models solve real software issues. This remote LATAM contract pays $25 per label and requires 20+ hours each week.

Review AI project tasks, outputs, verifiers, rubrics, and delivery quality as a remote QC Packaging Engineer. The 3-month contractor assignment requires 20 hours per week and English fluency.