Scientific Computing AI Agent Evaluator

About OpenTrain

OpenTrain AI is the hiring and contracting organization for this role. OpenTrain is the #1 platform for finding and building careers in AI training and data labeling, helping contributors discover projects, build a professional profile, and apply in minutes.

Creating an OpenTrain account is free, and your profile can help you present credible AI-training experience as you grow a long-term portfolio.

About AI Training Work

AI training is the human side of building artificial intelligence. Experts review model outputs, create examples, evaluate technical reasoning, and provide feedback that helps AI systems become more accurate and useful.

In this role, your scientific computing knowledge will help shape how AI coding agents approach complex technical problems, from ambiguous requirements and architectural updates to full-cycle engineering and research workflows.

The Role

OpenTrain AI is seeking a Scientific Computing AI Agent Evaluator to support AI training focused on scientific computing workflows, code sequences, and technical feedback. You will apply advanced STEM knowledge to evaluate and improve AI systems that assist with complex coding and technical work.

Prior AI training experience is not required when you bring significant professional or academic experience in scientific computing or another STEM discipline. This is an entry-level AI-training opportunity for contributors with substantial subject-matter expertise.

• Employment type: Part-time contractor
• Time commitment: 20+ hours per week
• Work location: Worldwide and fully remote
• Language: English
• Pay: $80–$100 per hour

What You'll Do

You will provide detailed examples of how AI coding agents can support advanced scientific computing workflows. Your work will combine hands-on technical execution with clear explanations and structured evaluation.

You will document technical approaches, challenges, and problem-solving strategies in written and verbal formats. You will also assess the practical impact of AI agents and provide feedback that can guide improvements to their capabilities.

• Demonstrate and document workflows using Codex, Claude Science, Claude Code, and Claude Cowork.
• Navigate large codebases, debug issues, and implement features with AI-agent support.
• Explain technical approaches, challenges, and problem-solving strategies clearly.
• Evaluate AI-agent performance against ambiguous requirements, architectural updates, and full-cycle technical scenarios.
• Assess tool strengths, limitations, and potential improvements.
• Contribute relevant data, scripts, or code sequences generated through agent-supported workflows.
• Review and validate AI-generated code, data, and other technical outputs.

Requirements

This role requires strong scientific computing or related STEM experience, along with the ability to judge whether AI-generated technical work is accurate, useful, and well-structured. You should be comfortable explaining intricate workflows to support high-quality AI training data.

Hands-on proficiency with AI tools is required. Experience using Codex, Claude Science, Claude Code, or Claude Cowork for advanced technical work is particularly relevant.

• Significant professional or academic experience in scientific computing or related STEM workflows.
• Hands-on experience integrating AI tools into advanced technical work.
• Ability to apply AI tools to complex engineering, research, or software challenges.
• Experience navigating large codebases, debugging issues, and implementing features with AI-agent support.
• Sound judgment when assessing and validating AI-generated code, data, and technical outputs.
• Strong written and verbal communication skills.
• Ability to document technical processes thoroughly and provide structured feedback.
• Careful attention to detail and a collaborative approach to improving AI systems.

Who Should Apply

Apply if you have substantial scientific computing, engineering, research, or related STEM experience and want to use that expertise in the fast-growing field of AI training. You do not need prior AI-training experience, but you should be ready to work hands-on with AI agents and critically evaluate their technical results.

This opportunity may suit professionals, researchers, academics, and technical contributors who can communicate complex concepts clearly while testing how AI systems perform in realistic scientific computing environments.

Why Work With OpenTrain

AI training and data-labeling work is a flexible way to contribute directly to cutting-edge AI development. Contributors help shape how modern AI systems reason, write code, and respond to demanding technical problems.

OpenTrain AI gives you a place to build a durable AI-training portfolio, highlight your expertise, discover relevant opportunities, and grow your career in this rapidly developing industry.