AI Trainer - Materials Science Specialists (Remote - Melbourne)

Role overview This is a remote advisory opportunity for experienced materials science professionals to support AI training and validation work on a flexible, project-by-project basis. Specialists will systematically review scientific content produced by large language models, comparing model-generated outputs against the original source papers to refine accuracy in the materials science domain. Joining the expert talent pool means participation is by invitation, and project availability depends on client demand and profile fit.

Responsibilities - Audit materials science research papers alongside LLM-generated graphical abstracts that summarize them - Identify specific cases where the AI-generated visual summary misrepresents the underlying paper - Provide detailed expert feedback and corrections to improve model accuracy - Complete paid evaluation tasks requiring up to one hour of concentrated focus per assignment - Contribute subject-matter expertise to enhance AI training datasets used by research teams - Maintain consistent quality standards across varied document types and complexity levels

Requirements - Master's degree or higher in materials science or a closely related discipline - Deep subject-matter expertise in materials science research, terminology, and methods - Prior experience with fact-checking, peer review, or AI-related evaluation work - A verifiable professional profile or LinkedIn presence demonstrating credentials - Strong attention to detail with the ability to concentrate on complex tasks for extended periods - Reliable high-speed internet connection and a personal computer suitable for remote work

Benefits and work setup - Fully remote position based in or around Melbourne, with flexible hours that fit around other commitments - Competitive hourly compensation typically starting at $40+/hour for qualified specialists - Short onboarding process once the qualification assessment is passed - Opportunity to shape next-generation AI models through applied domain expertise