About OpenTrain
OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We help people start and grow careers teaching AI by bringing opportunities together, building unified profiles, and making it easy to apply and track work across the industry.
Why AI training matters
AI training (also called data labeling or human feedback work) is the human side of building modern AI systems. Contributors annotate, evaluate, and refine examples that models learn from — work that’s remote, flexible, frequently entry-friendly, and directly shapes how state-of-the-art tools behave.
• Great part-time remote work: fit 20+ hours/week around other commitments.
• Accessible pathways: many projects require domain knowledge, attention to detail, or language fluency rather than formal credentials.
• Impactful: your feedback helps reduce hallucinations and improve agent reliability.
The role
OpenTrain is recruiting for an Agentic AI Coding Expert to work with autonomous coding agents on real engineering tasks. This contract, part-time engagement focuses on hands-on workflows: planning and executing multi-step coding tasks with AI assistance, evaluating AI-generated code for correctness and hallucinations, comparing outputs across systems, and documenting repeatable agentic development practices.
• Commitment: 20+ hours per week.
• Employment type: Contractor, Part-time.
• Location: United States only; English required (en).
What you'll do
You will run agentic development scenarios end to end, produce clear documentation of decisions, and deliver actionable feedback that improves models and tools.
• Use AI coding agents to plan, implement, debug, and refactor software.
• Execute multi-file edits, write and run tests, and verify behavior with agent assistance.
• Review AI-generated code to spot bugs, hallucinations, security issues, and maintainability problems.
• Compare code outputs and workflows across several AI systems and record reasoning and performance differences.
• Draft workflow guides and practical recommendations for reproducible agentic development environments.
• Provide detailed feedback on coding assistants, automation tools, and prompt patterns.
Requirements
Candidates must demonstrate hands-on experience using AI coding tools for multi-step technical tasks and strong written communication for documenting decisions and feedback.
• Experience delegating multi-step technical tasks to AI systems and refining results.
• Strong understanding of prompt engineering and AI collaboration/agentic workflows.
• Background in software engineering, DevOps, data science, ML engineering, or technical automation.
• Familiarity with AI coding tools such as Codex, GitHub Copilot, Claude Code, Cursor, or Gemini CLI.
• Excellent written and verbal English for documentation and feedback.
• Comfort comparing and reviewing multiple AI-powered development tools.
Helpful background
The following experience is useful but not strictly required to apply; they reflect the kinds of tasks you’ll perform and the contexts you may encounter.
• Practical experience with code generation, refactoring, and debugging.
• Test generation, code review, multi-file edits, terminal automation, and repository exploration.
• Familiarity with CI/CD, DevOps practices, or ML engineering workflows.
Compensation, logistics, and next steps
This is a paid contractor role. Hourly pay range is $70–$126 USD with a listed top rate of $126/hr. The project requires 20+ hours per week and is limited to candidates based in the United States. Work focuses on computer code/programming data and combines coding tasks with evaluation/rating deliverables.
• Data type: Computer code / programming; label types include coding and evaluation/rating.
• * To apply, prepare examples of prior work with AI coding tools, agentic workflows, or documentation of code reviews and send them per OpenTrain's application instructions.