AI Research Scientist, Learning & Evaluation

What you'll do

• Design and run model evaluations, including a multi-turn tutoring benchmark, and set release standards.
• Use product and expert-verified assessment data to inform model training and measure learning outcomes; build analytics and reporting across both products.
• Identify missing telemetry and work with engineering to improve instrumentation and establish measurement practices.

What they ask for

• PhD in a quantitative field plus 5+ years applying it to products or research, or a track record of owning evaluation or measurement for a shipped AI product.
• Production experience evaluating LLMs and the ability to explain an evaluation suite you built and its limitations.
• Rigor in experimental design, causal inference, controls, and sample-size reasoning.
• Clear written and verbal communication.

Nice to have

• Learning science, psychometrics, or item response theory.
• Experience with children's data and its associated rules.

Pay

$180,000–$280,000 USD base salary

Tools

Python, Pandas, NumPy, SciPy, Jupyter, SQL, MongoDB, PostgreSQL

AI focus

Evaluate and improve AI tutoring models for learning outcomes, including accuracy, reasoning, child safety, resistance to sycophancy, and Socratic teaching; build multi-turn tutoring benchmarks and turn product data into training signals.