JavaScript and TypeScript AI Model Evaluation Developer

You will combine full-stack JavaScript or TypeScript development with hands-on AI model training and evaluation. The work includes building code used to improve models, checking model responses against set criteria, and helping create training data.

You will also explain evaluation decisions, refine responses for clarity and technical accuracy, and work with researchers and annotators to improve model quality.

• Design, develop, and maintain readable, reusable, secure, and stable JavaScript or TypeScript code.
• Evaluate models, benchmark performance, review results, and recommend improvements.
• Rank responses to user questions across different subject areas using established criteria.
• Write clear explanations for your evaluation decisions.
• Create and maintain datasets for supervised fine-tuning.
• Support reinforcement learning with human feedback and reward-model improvement.
• Review code and documentation and provide useful feedback.
• Produce and refine model responses for clarity, relevance, and technical accuracy.

What It Pays and Takes

This is a remote, part-time contractor assignment with a one-month contract duration. The role is open worldwide, and you must be comfortable working in English.

• Pay: Not specified in the role details.
• Schedule: At least four hours per day and 20 hours per week.
• Weekly options: 20, 30, or 40 hours.
• Availability: Four hours of overlap with Pacific Time.
• Experience: The listing is marked entry level, while the role requires extensive JavaScript or TypeScript experience.
• Technical skills: Strong knowledge of JavaScript ES6 and at least one of Node.js, Nest.js, React, Angular, or Vue.
• Development skills: Experience building modular web applications with scalable architectures.
• Code quality: Ability to write organized, maintainable, correct, and clearly annotated code.
• Education: An engineering or computer science degree, or equivalent experience.
• Communication: Excellent spoken and written English.
• Valuable additional experience: Docker and software quality assurance or test planning.

How It Works

Apply on OpenTrain with your resume, then complete the application on the hiring site.

About AI Training Work

AI training is the human work behind systems that generate text and code. Contributors create examples, review model responses, rank outputs, and explain what makes an answer useful or accurate, while technical experience helps with specialized coding and evaluation tasks.