AI Neuroscientist — Benchmark Scientist, Remote

Latch; United States; Hybrid; USD 180,000 - 230,000; Full time

Job summary

Latch is seeking a scientific authority on ground truth for AI agents in neuroscience, building benchmark tasks and datasets across normal function and disease. You will define expert reasoning, write gold-standard answers, and curate multi-modal data bundles to enable rigorous evaluation of agent performance.

You will judge agent reasoning step by step, score confidence, and set hard calls on model systems and contested literature claims, with a culture that values radical transparency and

Qualifications

• Define expert reasoning, write gold-standard answers, and curate multi-modal data bundles.
• Set hard calls on model systems and contested literature claims.
• Judge agent reasoning step by step and score confidence.

Responsibilities

• Define expert reasoning, write gold-standard answers, and curate multi-modal data bundles.
• Set hard calls on model systems and contested literature claims.
• Judge agent reasoning step by step and score confidence.

Job description

Latch is seeking a scientific authority on ground truth for AI agents in neuroscience, building benchmark tasks and datasets across normal function and disease. You will define expert reasoning, write gold-standard answers, and curate multi-modal data bundles to enable rigorous evaluation of agent performance.

You will judge agent reasoning step by step, score confidence, and set hard calls on model systems and contested literature claims, with a culture that values radical transparency and