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

Role overview

A human data platform is engaging computer science researchers in the Birmingham area to evaluate AI-generated graphical abstracts of computer science papers. The position centres on quality assurance: ensuring that model-produced visual summaries accurately reflect the content and structure of original research. Work is fully remote, project-based, and offers flexibility around existing commitments.

Responsibilities

- Carefully read assigned computer science research papers and review their corresponding AI-generated graphical abstracts. - Cross-check whether the abstract conveys the paper's core contributions, methods, and findings without distortion. - Flag any factual inaccuracies, conceptual misstatements, or unsupported claims. - Evaluate the visual structure and logical flow of each abstract for coherence. - Provide reasoned judgements on whether technical terms are applied correctly within the summary. - Return timely evaluations through the platform's review interface.

Requirements

- A Master's or PhD in Computer Science (or a closely related discipline) is required. - Strong reading comprehension of peer-reviewed computer science publications. - Prior experience with fact-checking, peer review, technical editing, or similar analytical tasks. - High attention to detail when working with dense, technical material. - Ability to work independently and meet task deadlines remotely. - Based in or near Birmingham, with reliable connectivity for remote work.

Nice to have

- Familiarity with AI training data, annotation workflows, or model evaluation projects. - Experience in journal editing, scientific publishing, or programme committee work.

Benefits and work setup

- Fully remote engagement with flexible, ad hoc scheduling. - Competitive per-task compensation. - Quick onboarding once the initial qualification assessment is passed.