Research Engineer Neuroscience & AI

About the role

AI systems are becoming increasingly capable at scientific work, but we still lack rigorous ways to evaluate their understanding of biology—for example, whether they can interpret experimental data or recognize when a scientific conclusion is unsupported.

We're recruiting researchers to help build rigorous evaluations of AI capabilities in neuroscience.

Each evaluation presents an AI model with a scientific problem and measures its performance against a defensible reference. We're particularly interested in problems that require genuine scientific expertise—questions experienced researchers can answer but current AI systems struggle with.

This role combines computational research, experimental reasoning, and independent scientific problem-solving.

What you'll do

• Identify scientific questions across neuroscience that require practical research expertise.
• Find and interpret scientific literature and experimental datasets, including single-cell transcriptomics, high-content microscopy, electrophysiology, and genetic perturbation data.
• Translate biological questions into reproducible computational tasks with measurable outcomes and defensible reference answers.
• Run frontier AI models, investigate their predictions and reasoning, and identify successes and failures.
• Assess experimental limitations, confounders, alternative interpretations, data leakage, and grading errors.
• Develop, validate, and document research tasks, collaborating with scientists and engineers to improve their quality.

What we're looking for

• A PhD, completed or in progress, in neuroscience, neurobiology, computational biology, bioengineering, or a related field, or equivalent research experience.
• Depth in at least one area of neuroscience, such as functional genomics, in vivo assays, imaging, electrophysiology, neurodevelopment, or organoid biology.
• Experience analyzing experimental data computationally, with a working knowledge of experimental design and controls.
• Regular use of AI development tools such as Codex or Claude Code.

Nice to have

• Experience with Git and reproducible scientific analysis workflows.
• Experience with scRNA-seq analysis, perturbation experiments, high-content imaging, or neural activity measurements.
• Experience with human iPSC-derived neural systems, including differentiation, organoids, neuronal maturation, or disease modeling.

If you meet the core requirements, we encourage you to apply even without the nice-to-haves. Prior experience building AI evaluations is not required.

Compensation and engagement

This is a temporary, part-time W-2 research position paying $50–90 per hour , depending on experience.

The expected commitment is 10–20 hours per week for an initial two-month engagement, with the possibility of extension based on mutual interest and project needs.

Remote, hybrid, and onsite arrangements are available.

Selected researchers will receive paid onboarding. Applicable employee benefits and leave will be provided in accordance with employment arrangements and legal requirements.

Hiring process

• Screening interview
• Paid take-home research assignment
• Final interview and assignment review
• Offer

We aim to respond to applicants within two to four business days.

How to apply

Apply directly through LinkedIn.