OpenAI Cookbook Code Reviewer and Evaluator

About OpenTrain

OpenTrain AI is the hiring organization for this role. OpenTrain is the #1 platform for building careers in AI training and data labeling — a place where contributors start and grow work that directly shapes how modern AI systems behave.

About AI Training Work

AI training (data labeling/annotation) is the human side of building intelligent systems: people prepare, review, and critique examples that models learn from. These roles are commonly remote, flexible, and accessible, and contributors work on tasks such as reviewing model outputs, annotating code, and providing high-quality feedback that improves model behavior.

• Work is remote and flexible — fits part-time schedules.
• Many projects require no formal degree; domain expertise is paid where needed.
• Contributors directly influence model accuracy, safety, and usefulness.

Role Overview

We are hiring a contractor to evaluate AI-generated code and explanations related to OpenAI Cookbook implementations. You will run structured AI-driven interviews with candidates, review generated API code and explanations, and provide precise, actionable labeling and feedback aligned to Cookbook best practices.

• Position type: Contractor, part-time (less than 20 hours/week).
• Pay: $20 USD per hour.
• Worldwide applicants accepted; work is remote.

What You’ll Do

This role combines technical code review with structured evaluation and labeling. You will assess correctness, efficiency, and adherence to documented best practices, then record labels and write structured feedback that helps improve future model outputs.

• Analyze AI-generated prompts, API calls, code snippets, and explanations for accuracy and efficiency.
• Label and categorize responses according to provided criteria and taxonomy.
• Identify errors, inefficiencies, and missing optimizations; suggest concrete improvements.
• Conduct technical interview interactions following supplied guidelines to validate candidates’ hands-on experience.
• Provide concise, well-structured written feedback in clear English.

Key Requirements

Candidates must preserve the stated requirements exactly: this role is technical and requires demonstrated hands-on experience and strong communication.

• 5+ years hands-on experience working with OpenAI’s API and Cookbook best practices.
• Proven experience with fine-tuning, embeddings, tokenization strategies, prompt engineering, and API optimizations.
• Experience deploying GPT-based applications or similar production work (chatbots, assistants, document processing, etc.).
• Strong English writing skills — ability to produce structured, concise feedback.
• Comfort evaluating AI-generated code, debugging snippets, and explaining fixes in plain language.

Interview & Evaluation Tasks You’ll Run

You will follow the supplied interview guidelines to probe candidates’ depth of knowledge. Tasks include presenting buggy API snippets, asking for optimizations, evaluating claims about fine-tuning vs embeddings, and testing the candidate’s ability to explain tokenization simply.

• Present and debug example API code (e.g., incorrect max_tokens usage) and explain fixes and cost/quality trade-offs.
• Assess candidate answers for concrete, hands-on examples versus theoretical knowledge.
• Score AI-generated explanations (for example, correcting an incorrect statement about fine-tuning GPT-4) and annotate what’s wrong and how to fix it.
• Ensure feedback follows the project’s structured format and labeling schema.

Who Should Apply

Apply if you are an experienced developer with deep practical knowledge of OpenAI API workflows and Cookbook techniques, enjoy technical review and mentoring, and can communicate complex concepts clearly in English. This is a good fit for engineers who like short, focused contract work and precise evaluation tasks.

• Ideal for software engineers, ML engineers, or technical leads with real-world API experience.
• Not for applicants who only have theoretical familiarity — hands-on debugging history is required.
• Entry on the platform will reflect contractor, part-time availability.

Logistics & How To Apply

This is a contract, part-time role paid at $20/hour with expected commitment of less than 20 hours per week. You will receive interview scripts, evaluation rubrics, and labeling templates to follow. Apply with examples of past OpenAI API projects, links or descriptions of work where you applied Cookbook techniques, and confirmation of availability.

• Payment type: PAY_PER_HOUR in USD at $20/hour.
• Hours: Less than 20 hours/week; worldwide applicants welcome.
• Employment types: Contractor, Part-time.