IT/SaaS Ops Expert — AI Data Trainer (Contract)

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 profiles, and apply in minutes so they can start shaping how AI behaves.

We connect experienced people to short-term, remote projects across the AI training lifecycle. This role is an opportunity to join that community and contribute your SaaS operations knowledge to real-world AI systems.

Why AI Training Work Matters

AI training (data labeling / human feedback) is the human side of building intelligent systems. Modern models learn from the examples and judgments that people provide — from labeling text to reviewing model responses — and this work directly influences how AI agents behave.

These projects are 100% remote, often flexible and part-time, and accessible to professionals who bring domain knowledge, careful reasoning, and clear writing rather than advanced coding skills.

The Role — IT/SaaS Ops Expert (AI Data Trainer)

You will review structured B2B SaaS workplace scenarios and label whether specific information is appropriate for an AI agent to use and whether it should be visible to different people in the scenario.

Labels are simple: yes / no / undecidable plus a short, defensible justification. Work is remote, hourly contract, roughly 20 hours per week for about 2.5 months, paid at $40 USD per hour.

• Contract type: Contractor, Part-time
• Estimated hours: ~20+ hours/week
• Duration: ~2.5 months
• Pay: $40 USD per hour
• Data type: Text; Label types: Classification, Evaluation/Rating, Question Answering; Tooling: OTHER

What You'll Do Day-to-Day

Apply a structured rubric to short workplace scenarios and decide whether particular details are appropriate for an AI agent to use and who in the scenario should see them. Provide concise, defensible justifications for each judgment.

Work independently on labeled tasks, read brief technical context (ticket snippets, access logs, onboarding notes), and make consistent decisions even when scenarios are ambiguous.

• Read scenario descriptions and related context and mark visibility/appropriateness labels (yes / no / undecidable).
• Write short justifications that explain your reasoning clearly in English.
• Follow rubric rules to keep labeling consistent across similar cases.
• Flag ambiguous or novel scenarios for reviewer guidance when needed.

Requirements

Candidates must meet all listed requirements; we will preserve these standards to ensure high-quality, consistent judgments for the training data.

You do not need to be an expert coder, but you must be comfortable reading basic technical context and making defensible judgments.

• Experience: 1+ year across IT support, security operations, SRE, customer success, or similar SaaS operations work at a B2B SaaS company.
• Education: Bachelor’s degree or equivalent on-the-job IT/SaaS experience.
• Domain expertise: provisioning, escalations, incident handling, onboarding/offboarding, access reviews across SaaS Ops sub-domains.
• Tools: Familiarity with SaaS Ops tools (ticketing systems, IAM, monitoring/alerting, CRM).
• Skills: Strong English writing and careful reasoning; high attention to detail; ability to apply structured rubrics consistently.
• Role fit: Cross-domain breadth (generalist) rather than a narrow specialty.

Who Should Apply

Apply if you have hands-on experience in SaaS operations or adjacent IT roles and enjoy concise technical writing and judgment work.

This project is well suited to professionals who want part-time remote work that leverages their operational experience to shape AI behavior.

• Good fit: IT support agents, SREs, security operations analysts, customer success managers with technical SaaS experience.
• Not required: advanced coding skills or deep software engineering expertise.
• Eligible countries: US, CA, GB, IE, AU. English fluency required.

How It Works — Process, Tools & Next Steps

You will complete labeling tasks inside a provided annotation interface (labeled as OTHER in the brief). Each task includes the scenario, any relevant context, the rubric, and fields for your yes/no/undecidable decision plus a brief justification.

Expect a training/check phase to align on the rubric, followed by production tasks. Work schedule is flexible; you can choose hours that fit your availability while meeting throughput expectations.

• Annotation type: Text classification, evaluation ratings, and short question-answer judgments.
• Onboarding: brief training and calibration tasks to ensure rubric alignment.
• Communication: project reviewers will provide feedback on borderline judgments during the initial phase.