Senior Bash Code Reviewer

About OpenTrain

OpenTrain AI is the hiring and contracting organization for this role. OpenTrain is the #1 platform for finding and building careers in AI training and data labeling, helping contributors discover cutting-edge projects, build experience, and apply in minutes.

• Part-time contractor opportunity
• Remote work for eligible contributors in India, the Philippines, or Vietnam
• Pay: $21 USD per hour

About AI Training Work

AI training is the human side of building modern artificial intelligence. Skilled reviewers evaluate examples, including code and model responses, so AI systems can become more accurate, reliable, secure, and useful.

In this role, your Bash expertise will directly support the quality of data used to train next-generation AI models. AI training work is remote and flexible, making it possible to contribute to advanced technology alongside other commitments.

• Help improve how AI systems generate and evaluate shell scripts
• Apply professional technical judgment to model-generated code
• Contribute to a fast-growing field at the intersection of software and AI

The Role

OpenTrain is hiring a Senior Bash Code Reviewer to audit annotator reviews of AI-generated shell scripts. You will validate each script in a sandboxed environment, confirm that it satisfies the prompt, and assess whether it runs correctly across POSIX-compliant shells.

You will also check security and reliability practices, correct mis-ratings, and provide clear, actionable feedback so every review meets the project quality rubric.

• Time requirement: 20+ hours per week
• Language: English
• Experience level listed for the project: Intermediate
• Core focus: AI-generated Bash code response annotation

What You’ll Do

Your work will combine hands-on shell execution, technical code review, and structured quality assurance. You will assess both the behavior of scripts and the quality of the original annotator review.

• Execute Bash and shell scripts in isolated virtual machines, containers, or other sandboxed environments
• Verify that scripts satisfy the original prompt and run correctly across POSIX-compliant shells
• Identify security issues such as unescaped variables, injection risks, insecure file handling, unsafe eval, and unsafe sourcing
• Review input sanitization, proper quoting, least-privilege practices, and environment configuration
• Assess error handling, exit codes, traps, pipelines, set -euo pipefail usage, and idempotent behavior
• Correct inaccurate ratings and explain the appropriate rating using the quality rubric
• Provide concise, constructive, actionable feedback to annotators
• Use checklist reviews, rubric-based scoring, and ticketing or annotation tools

Required Qualifications

This role is designed for an experienced Bash engineer, Linux systems administrator, or dedicated code reviewer who can make consistent technical judgments about shell code and communicate them clearly in written English.

• 5 to 7 or more years of professional Bash scripting, Linux systems administration, or code-review experience
• Expert knowledge of Bash 4.x and 5.x, POSIX sh compatibility, subshells, pipelines, traps, and environment configuration
• Strong ability to detect shell injection risks, unescaped variables, insecure file handling, and unsafe eval or sourcing
• Experience with robust error handling, reliable exit codes, and re-runnable, idempotent logic
• Proficiency with Bats, ShellCheck, and bash -x tracing
• Familiarity with cron jobs, systemd units, and containerized Linux environments
• Comfort validating proof of work by executing scripts in isolated VMs or containers
• Experience with rubric-based scoring, checklist reviews, and Jira, Asana, or similar tools
• Excellent written English at B2 CEFR level or higher

Communication and Collaboration

The strongest reviewers combine technical precision with clear communication. You should be able to explain why a script or rating is incorrect, provide constructive feedback, and mentor peers when needed.

• Communicate technical findings clearly in written English
• Give specific, actionable feedback rather than unsupported judgments
• Maintain consistency when applying a review rubric
• Support peer learning through constructive guidance and mentoring

Nice-to-Have Experience

Experience in AI evaluation or data-labeling environments is helpful, although the core requirement is strong Bash and Linux expertise.

• Exposure to large language model evaluation
• Experience with RLHF pipelines
• Prior AI or machine-learning data-labeling work

Contract Details and Application

This is a part-time contractor opportunity paying $21 USD per hour, with a commitment of at least 20 hours per week. The role is available to contributors located in India, the Philippines, and Vietnam.

Create a free OpenTrain account to build your profile and apply through OpenTrain. Your Bash, Linux, testing, security, and review experience can help shape how AI systems handle real-world shell scripting tasks.

• Employment type: Contractor
• Schedule: Part time, 20+ hours per week
• Rate: $21 USD per hour
• Eligible countries: India, Philippines, Vietnam