AI Trainer – Computer Science Expert - Graphical Abstract - Fully Remote - (Montreal)

Role overview

A remote opportunity is open for computer science specialists to review and verify AI-generated graphical abstracts of scientific research papers. The position contributes to a project that benchmarks how accurately large language models summarise dense technical literature, with the reviewer judging both factual fidelity and structural integrity. Work is fully remote, ad hoc, and offers flexible hours suited to academics who can take on tasks around other commitments.

Responsibilities

- Read peer-reviewed computer science papers and compare them side-by-side with LLM-generated graphical abstracts - Identify factual, conceptual, or visual errors where the abstract misrepresents the source material - Assess whether key technical methods, results, and terminology are captured correctly in text and imagery - Provide clear written feedback flagging specific issues for the project team - Work independently through assigned tasks on a flexible schedule

Requirements

- Master's or PhD in Computer Science or a closely related STEM discipline - Demonstrated ability to navigate and critically evaluate dense scientific literature - Strong analytical reasoning and fact-checking skills - High attention to detail and consistent judgement across tasks - Comfortable working fully remote with self-directed scheduling

Nice to have

- Prior experience with AI training, annotation, or model evaluation work - Background in scientific publishing, peer review, or technical editing

Benefits and work setup

- Fully remote, work-from-home arrangement (Montreal-based applicants welcome) - Ad hoc, task-based engagements with flexible hours - Competitive per-task pay rates - Quick onboarding once an assessment is passed