Senior JavaScript (React) Code Reviewer

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. Contributors use OpenTrain to discover projects, build a profile, and apply quickly; creating an account is free.

We focus on the human side of building AI — people who prepare, review, and improve the examples that modern models learn from. If you want to apply software engineering expertise to shape how AI behaves, this is where that work happens.

About AI training & data-labeling work

AI training (also called data labeling or human feedback work) is how human expertise is turned into reliable training data for machine learning. Annotators and reviewers evaluate model outputs, verify correctness, and flag security or quality issues so models learn from high-quality examples.

These projects are often 100% remote and flexible, making them a great way to contribute your skills part-time while staying at the cutting edge of AI development.

The role

You will audit evaluations produced by annotators who assessed AI-generated JavaScript and React code. Your job is to run proof-of-work checks, confirm that each snippet follows the prompt, executes correctly in a sandboxed environment, and meets security and guideline standards.

This is a contractor, part-time role requiring 20+ hours per week. Pay is $24 USD per hour. The position is remote and open worldwide.

• Employment: Contractor, Part-time
• Workload: 20+ hours/week
• Pay: $24 USD per hour
• Location: Remote, worldwide

What you'll do day to day

You will review annotator ratings and written feedback, verify functional correctness, and catch rating errors or guideline violations. When necessary, you'll execute code in a sandbox or container to reproduce behavior and validate results.

Your feedback should be concise, actionable, and aligned with the project rubric to protect the accuracy of training data.

• Audit annotator evaluations of AI-generated JavaScript/React responses.
• Run proof-of-work checks via sandbox/container execution to verify prompt compliance and correctness.
• Confirm security best practices and flag XSS, CSRF, injection, supply-chain, or permission issues.
• Assess tests and edge cases, review Jest/RTL/Cypress or Playwright output when provided.
• Provide concise, rubric-aligned feedback and escalate ambiguous cases per project guidelines.

Requirements

All role facts below come from the project requirements; please ensure you meet them before applying.

• 7+ years of professional JavaScript and React development, QA, or dedicated code-review experience.
• Deep knowledge of modern React: Hooks, Context, Suspense, Concurrent features, SSR/Next.js, and common state-management patterns (Redux, Zustand, Context + Reducer).
• Fluent in modern JavaScript (ES2023+) and TypeScript; experience with module bundlers (Vite/webpack) and package managers (npm, pnpm, Yarn).
• Advanced testing & debugging skills with Jest, React Testing Library, Cypress or Playwright; experience with coverage analysis and profiling.
• Proven ability to detect and mitigate common web vulnerabilities (XSS, CSRF, injection, supply-chain, permission issues).
• Experience optimizing performance (bundle size, Web Vitals, rendering) and ensuring accessibility (WCAG-compliant UI code).
• Comfortable with Docker (build/run/exec), CI/CD pipelines (GitHub Actions, GitLab CI), and code-review platforms (GitHub PRs, GitLab MRs).
• Hands-on with sandbox/container execution to validate functional correctness and prompt compliance.
• Familiarity with rubric-based scoring, ticketing tools (Jira, Asana), and structured annotation workflows.
• Excellent written English (B2+ CEFR) for concise, constructive feedback and documentation.

Nice to have

These additional experiences are helpful but not required for all projects.

• Exposure to LLM evaluation, RLHF pipelines, or data-labeling platforms such as Labelbox or Scale AI.

Who should apply

This role is a strong fit for senior JavaScript/React engineers, security-focused front-end developers, and experienced code reviewers who want flexible, remote work that directly improves AI training data.

If you enjoy debugging, security reviews, and turning detailed technical judgment into clear feedback, you’ll thrive here.

How the work typically flows

You will receive batches of annotator evaluations to audit. For each item you will confirm the annotator's rating against the rubric, reproduce behavior in a sandbox when needed, and add concise feedback or corrections. Follow project-specific guidelines and escalate uncertain cases.

Communication and documentation are important: use the project’s ticketing or review channels (Jira, GitHub, etc.) to note recurring issues or gaps in the rubric.