AI Trainer - Neuroscience; Fully Remote- Vancouver

Role overview

This remote advisory role recruits neuroscience specialists to judge the quality of graphical abstracts that AI systems have produced from scientific papers. The aim is to ensure machine-generated technical summaries preserve the accuracy and nuance of the original neuroscience research. The engagement is structured as a flexible, ad hoc project, making it well suited to researchers seeking supplemental, expertise-driven work.

• Read assigned neuroscience papers and evaluate the corresponding LLM-generated graphical abstract for fidelity.
• Identify where visuals or captions misrepresent methods, results, or conclusions.
• Confirm that technical vocabulary, statistical claims, and scientific concepts are rendered correctly.
• Produce structured feedback describing errors and the source passages they contradict.
• Pass a paid skills assessment using a sample paper and abstract to enter the participant pool.

• Master's or PhD in neuroscience or a closely related field.
• Strong grasp of experimental neuroscience, including common methodologies and analytical approaches.
• Track record of careful, evidence-based review of scientific or technical documents.
• Sharp analytical eye and ability to articulate findings in concise written form.
• Comfort operating independently within a project-based remote framework.

• Prior work on AI training, evaluation, or human-in-the-loop annotation tasks.
• Experience in scientific publishing, manuscript review, or editorial decision-making.

• Fully remote position open to candidates based in the Vancouver area.
• Flexible scheduling with ad hoc assignments and hourly compensation starting at $40+.