Real Claude Code CLI Execution Traces Contributor

About OpenTrain

OpenTrain AI is the #1 platform for finding and building careers in AI training and data labeling. We connect contributors with opportunities to help shape the systems behind modern artificial intelligence, while making it easy to build experience and apply to projects.

For this role, OpenTrain AI is hiring and contracting expert software engineers to produce high-quality coding-agent training data.

About AI Training Work

AI training is the human side of building artificial intelligence. People prepare examples, review model behavior, and create feedback that helps AI systems become more capable, accurate, and useful.

This project focuses on software engineering trajectories: authentic records of how people use coding agents to navigate repositories, develop features, debug issues, refactor code, run tests, and validate results.

The Role

As a Real Claude Code CLI Execution Traces Contributor, you will submit authentic supervised fine-tuning trajectory data from real software development sessions. Your submissions will document complete interactions between a professional engineer and the Claude Code CLI across simple, medium, and complex programming tasks.

This is an expert-level contractor opportunity for daily Claude Code CLI users with at least two years of professional software engineering experience. The expected project duration is 6+ months.

• Work remotely as a contractor with flexible contribution scheduling.
• Compensation is listed at $3 per verified trace.
• Contribute across simple, medium, and complex task tiers.
• Discuss your relevant experience with software engineering and AI coding agents during screening.

What You'll Do

You will create and submit verified JSON or JSONL execution traces from genuine software development work. Each trace must preserve sufficient context for reliable review and model training, including the starting state of the repository, the development process, tool interactions, and validation evidence.

• Use the Claude Code CLI during authentic software engineering sessions.
• Capture the exact starting repository commit.
• Include root-level project guidelines and complete terminal and tool outputs.
• Preserve intermediate reasoning and the full session context.
• Cover repository navigation, debugging, feature development, refactoring, testing, and terminal workflows.
• Submit proof such as passing tests and a clean git diff.
• Deliver an initial batch of 20–50 traces.
• Maintain weekly throughput across the required complexity tiers.
• Provide sample logs when available.
• Ensure each trace is accurate, complete, and reliably validated.

Requirements

This project requires substantial professional software development experience and practical familiarity with AI coding agents. Contributors must be able to explain their decisions clearly and preserve trustworthy evidence from each session.

• Daily user of the Claude Code CLI.
• At least two years of professional software engineering experience.
• Practical expertise across common programming stacks.
• Experience with repository navigation, debugging, feature development, refactoring, and testing.
• Strong command of terminal workflows and authentic human-agent interaction.
• Ability to preserve complete session context and reliable validation evidence.
• Relevant software engineering and AI coding agent experience.
• Strong accuracy and attention to detail.
• Ability to explain screening decisions clearly.
• Fluent English proficiency.

Who Should Apply

This opportunity is designed for expert software engineers who regularly work in the terminal and understand how professional development tasks unfold in real repositories. It is a strong fit for contributors who can produce detailed, authentic coding-agent records rather than simplified demonstrations.

Applicants should be comfortable discussing their experience with Claude Code CLI workflows, software engineering practices, and AI coding agents. The project is available worldwide.

Project Deliverables And Quality

Every submission should represent a complete and reviewable development trajectory. Quality depends on preserving the original context, recording the interaction faithfully, and providing evidence that the resulting work was validated.

Your traces should be delivered in JSON or JSONL format and organized so reviewers can verify the repository state, actions taken, outputs produced, and final result.

• Initial delivery: 20–50 execution traces.
• Ongoing weekly throughput by complexity tier.
• Pricing based on each verified trace.
• Complete session context and supporting logs.
• Validation evidence, including passing tests and a clean git diff when applicable.

How To Get Started

Create a free OpenTrain account to build your profile and apply in minutes. During screening, be prepared to discuss your relevant software engineering and AI coding agent experience, your Claude Code CLI usage, and your ability to produce accurate, fully validated execution traces.

• Apply as a contractor through OpenTrain AI.
• Share relevant experience for Real Claude Code CLI Execution Traces.
• Complete the project screening process.
• Begin contributing verified traces if selected.