Product Data Scientist

About the role

No scalpel. No blood. No hospital bunker. The ZAP-X treats tumors and other conditions deep in the brain with a self-shielding gyroscopic robot that steers radiation around the head to sub-millimeter precision — built to put world-class radiosurgery within reach of patients everywhere, not just a handful of elite centers. You’ll join our Product team as a hands-on data scientist, working closely with the Director of Product Management to turn data from our global fleet — treatment plans, delivery and imaging data, system telemetry, and quality records — into models, analyses, and insights that shape what we build. It’s a build-heavy role with room to grow: real modeling problems, direct mentorship, and a front-row seat to how AI is reshaping radiosurgery. You’ll work across software, clinical, quality, and service teams.

What you’ll do

• Shape the future of AI-assisted treatment planning and delivery. Working closely with the Director of Product Management, you’ll evaluate how AI capabilities — built in-house or brought in through partners — perform on real cases, building the baselines and benchmarks behind the call and helping define what “good enough” looks like. Engineering builds and runs the production systems; you do the hands-on analysis and modeling behind the direction.
• Tell the story in our data. Use statistical and causal modeling to turn treatment, outcomes, and fleet data into insights that inform leadership’s decisions — how a new feature changes treatment efficiency or patient outcomes across sites, and how the product is actually used, to help sharpen what we build next.
• Catch problems early. Build predictive and anomaly-detection models, and the early-warning signals on top of them, that surface reliability and quality issues — and support the quality team’s complaint trending and post-market surveillance — before they escalate into complaints or downtime.
• Make data self-serve. Build dashboards and AI-assisted tooling so quality, operations, and service teams can investigate issues and answer routine questions for themselves.

Your first 12 months

By the end of your first year, you’ll have:

• Ramped fast — leaning on AI tools to get up to speed on ZAP’s data and domain in weeks, not months — and shipped your first models into use.
• Put predictive and anomaly-detection models to work flagging reliability and quality issues early for the service and quality teams.
• Delivered analyses across product, clinical, and fleet questions that leadership acted on.
• Helped evaluate at least one AI planning or delivery capability, building the baselines and benchmarks behind the recommendation.
• Built AI-assisted self-serve tooling that lets other teams query the data directly, not just read static dashboards.
• Become a go-to person for a meaningful slice of ZAP’s data, on a clear path to broader scope.

What we’re looking for (the essentials)

Typically, around 3 years of data science experience.

• A Bachelor’s or Master’s in a quantitative field — computer science, statistics, engineering, physics, or similar — or equivalent hands-on experience.
• Proficient in Python, with a solid grounding in statistics and core ML methods.
• Hands-on experience building models or analyses end to end — from messy data to a result someone actually used.
• Clear communication — you can explain a finding to a non-technical audience.
• Eager to learn the domain and grow into bigger problems.

You don’t need to tick every box below. If you’re strong on the essentials and excited by the mission, we’d like to hear from you.

Nice to have

• Healthcare, medical-device, or other regulated-industry experience.
• Experience with time-series, anomaly detection, or causal inference.
• Familiarity with imaging or computer-vision ML.
• Comfortable querying data with SQL.
• Exposure to dashboarding or analytics tooling.

This role could suit someone from a data science, ML, or analytics background who wants to build and grow — we care more about how you think with data than your exact title today.