Freelance Academic Research Expert AI Evaluator

Uber AI Solutions is Uber's marketplace connecting skilled freelancers with Generative AI researchers and product teams working on cutting-edge AI systems. We partner with independent contractors to support large-scale model evaluation, alignment, and quality initiatives.

As a Freelance Academic Research Expert, you will pair genuine scholarly expertise with hands-on GenAI-evaluation experience to design challenging, realistic prompts backed by source documents such as peer-reviewed papers, published datasets, methodology standards, and systematic review protocols. You will then probe where the model breaks, and encode expert judgment into precise, gradable rubrics and a canonical 'golden' answer. You own an end-to-end evaluation task from prompt design through reviewer sign-off.

This is hands-on academic work with an analytical core. You will produce scholarly deliverables as the output; what makes the work difficult is evaluating methodology, evidence, and citation integrity against the provided sources. It is not a teaching, curriculum delivery, or tutoring role.

What you'll work on

• Apply your academic research expertise to help create and evaluate complex AI prompts and responses.
• Conceptualize and draft realistic, multi-step prompts that mirror real research assignments and require synthesis, methodological evaluation, and reconciliation across multiple sources.
• Identify and document genuine model failures by writing specific, objective, and actionable explanations.
• Build comprehensive evaluation rubrics that trace each required input through dependent criteria and carefully separate extraction from interpretation.
• Produce fully correct, deliverable-quality golden responses that satisfy all criteria using discipline-standard conventions and citation practice.
• Complete our onboarding process and strictly follow the disciplined authoring and review workflow for each task.
• Closely follow guidelines, run AI quality checks on your own work, and resolve any issues in the feedback loop before final submission.

Engagement details

• Location: Remote (United States)
• Engagement Type: 1099 Freelance / Independent Contractor, through the Uber AI Solutions platform
• Engagement Duration: 40 hours over 1-2 weeks

Who we're looking for

• Ph.D., Ed.D., or completed Postdoctoral fellowship from an accredited institution.
• 5+ years of experience in academic research, peer review, or higher education instruction.
• Demonstrated ability to read primary source documents (peer-reviewed literature, published datasets, methodology standards) and derive multi-step analytical conclusions from them.
• Familiar with research methodology evaluation, systematic literature review, and citation integrity standards.
• Comfortable working independently on detail-oriented tasks.
• Strong analytical and written communication skills.
• Prior hands-on experience with GenAI / LLM evaluation and prompt engineering, rubric or golden-answer authoring, data annotation, RLHF, red-teaming, or model quality assessment.

Why this matters

Your expertise will help improve how AI systems handle complex scholarly topics, making outputs more accurate, reliable, and better aligned with real-world research reasoning.