Senior TypeScript (React) Code Reviewer

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect people to projects that teach AI systems how to behave and provide tools to build a career reviewing and annotating real model outputs.

Working with OpenTrain means joining a fast-growing industry where contributors directly shape state-of-the-art AI systems. Many roles are remote and flexible, making it easy to fit this work around other commitments.

About AI training and this work

AI training (data labeling/annotation) is the human side of building machine learning systems. For generative models, that often means evaluating code outputs, correcting mistakes, and ensuring safe, performant, and accessible behavior.

This project is focused on evaluating AI-generated TypeScript and React code — a high-impact task because accurate reviews and feedback directly improve models used by developers worldwide.

The role

We’re hiring a senior TypeScript engineer with deep, production-level React expertise to audit annotator reviews of AI-generated TypeScript + React code responses.

You will install dependencies, compile with tsc, execute snippets in a sandboxed container, and verify that each submission meets the prompt, renders correctly, and follows best practices for security, performance, and accessibility.

What you'll do

Deliver consistent, rubric-aligned reviews and correct any mis-ratings with clear, constructive feedback to annotators. Protect the dataset quality used to train next-generation models.

• Install dependencies and build submissions (tsc) to confirm compile-time correctness and strict-mode compliance.
• Run code in containerized/sandboxed environments to verify runtime behavior, rendering, and prompt compliance.
• Assess security risks (XSS, CSRF, injection, supply-chain CVEs) and flag or correct unsafe patterns.
• Evaluate performance considerations: bundle size, code-splitting, lazy loading, and impact on Web Vitals.
• Check accessibility against WCAG principles and recommend fixes for problematic components.
• Correct annotator ratings when they deviate from the rubric and provide concise, actionable feedback.
• Use testing and profiling tools (unit/e2e test results, coverage, React DevTools, performance profilers) to support decisions.

Required experience and skills

Every requirement below comes from the project brief — applications that match these will perform best in this role.

• 7+ years professional TypeScript/React development, QA, or dedicated code-review experience.
• Expert with React 18+: Hooks, Suspense, Context, and Server Components plus common state-management patterns (Redux, Zustand, Jotai).
• Advanced TypeScript: generics, utility types, strict configs, decorators, and modern ES2023+ features.
• Proven testing and debugging skills with Jest, React Testing Library, and Cypress or Playwright; familiarity with coverage tools (Istanbul/nyc).
• Secure coding expertise: detect/mitigate XSS, CSRF, injection, and supply-chain CVEs; familiarity with scanners like Snyk or npm-audit.
• Performance & accessibility experience: bundle optimization, code-splitting, lazy loading, CLS reduction, and WCAG compliance.
• Toolchain proficiency: Vite/webpack, pnpm/Yarn, Docker, and CI/CD (GitHub Actions, GitLab) workflows.
• Comfort executing code in containerized or sandboxed environments to validate functionality and prompt compliance.
• Structured QA practice: rubric-based scoring, checklist reviews, and experience with ticketing/annotation tools (Jira, Asana).
• Excellent written English (B2+ CEFR) to deliver concise, constructive feedback and mentor peers.

Nice to have

• Prior experience evaluating LLM outputs, RLHF pipelines, or other AI/ML data-labeling projects.
• Familiarity with code-review platforms and PR/MR workflows at scale.

Who should apply

Experienced TypeScript/React engineers who enjoy code review, mentoring, and improving data quality for AI models should apply. This role suits someone comfortable running and debugging code in containers, enforcing security and accessibility best practices, and writing clear corrective feedback.

How the work is arranged

This is a remote contractor, part-time role (20+ hours/week). Work is paid hourly at USD $25/hr. Open to applicants worldwide.

You'll receive submissions through a labeling platform (custom/OTHER) and use your own development environment plus container tooling to validate code. Reviews follow a supplied rubric; mis-ratings must be corrected with clear comments and remediation guidance.

• Employment type: Contractor, Part-time.
• Time requirement: 20+ hours per week (flexible scheduling).
• Pay: $25 USD per hour, paid per contractor agreement.