AI Trainer - Neuroscience; Fully Remote- UK

Role overview

This remote advisory position engages neuroscience specialists to evaluate graphical abstracts that large language models have produced from peer-reviewed research papers. The contribution helps safeguard scientific accuracy in machine-generated summaries used across AI development. Work is structured as a flexible, ad hoc project suitable for researchers who want to influence how AI represents complex scientific findings.

• Review neuroscience publications alongside their LLM-generated graphical abstracts.
• Detect inaccuracies, omissions, or visual mischaracterisations in the generated abstracts.
• Confirm that the graphical summaries faithfully reflect the experimental design, results, and conclusions of the original paper.
• Document specific errors and produce structured critiques for the project team.
• Complete a paid skills assessment using a sample paper and abstract before joining the participant pool.

• Master's or PhD qualification in neuroscience or an adjacent field.
• Substantial expertise in reading and interpreting primary neuroscience research.
• Experience with detailed fact-checking or structured scientific review.
• Ability to articulate technical critique clearly and concisely.
• Comfort working independently in a remote, asynchronous setting.

• Previous exposure to AI training, annotation, or evaluation workflows.
• Experience as a peer reviewer, editor, or contributor to scientific journals.

• Fully remote position open to UK-based participants.
• Flexible scheduling with ad hoc assignments and competitive pay starting at $40+/hour.