Senior Machine Learning Data Scientist

Our mission at Oura is to empower every person to own their inner potential. Our award-winning products help our global community gain a deeper knowledge of their readiness, activity, and sleep quality by using their Oura Ring and its connected app. We’ve helped millions of people understand and improve their health by providing daily insights and practical steps to inspire healthy lifestyles.

We are looking for a Senior Machine Learning Data Scientist to join Ōura’s Health Foundation Models team and develop foundation models of longitudinal wearable data that unlock new health insights and product experiences. We are looking for a researcher-builder who is excited to push the frontier of what these models can do—and hungry to turn those advances into capabilities that reach Ōura members. You will help shape an evolving research agenda rather than own a narrowly predefined problem, taking ambitious ideas end to end: from identifying the right question and developing new modeling approaches, through building the systems needed to test them and rigorously evaluating the results, to prototyping and shipping real-world applications.

What You Will Do:

• Shape and lead ambitious research directions for foundation models of longitudinal wearable and physiological data.
• Own research end to end: turn open-ended questions into testable hypotheses, develop new modeling approaches, and build the datasets and experimental infrastructure needed to investigate them.
• Establish rigorous evaluation methods that distinguish meaningful model capabilities from results driven by leakage, confounding, or fragile benchmarks.
• Advance capabilities such as representation learning, forecasting, personalization, multimodal modeling, and generative modeling across large longitudinal datasets.
• Drive promising model capabilities from research result to prototype, validation, and shipped health experiences for Ōura members.
• Share scientifically important advances through publication when the work warrants it, without losing focus on product impact.
• Collaborate with scientists, clinicians, engineers, and product partners while independently driving work through ambiguity.

This is a remote role within the United States, with a slight preference for candidates based in New York or elsewhere on the East Coast. There is no in-office requirement.

What you'll bring

• 5+ years of relevant machine learning research and applied experience, including experience gained during doctoral research, and a PhD or MSc in machine learning, computer science, statistics, electrical engineering, or a related quantitative field.
• Exceptional research judgment: a track record of identifying important questions, forming original hypotheses, and designing experiments that produce credible evidence.
• Deep technical expertise in modern foundation models, representation learning, large-scale training, and model evaluation.
• Demonstrated autonomy in taking ambiguous research problems from initial idea to a working system and clear scientific result.
• Strong grounding in probability, statistics, experimental design, and robust evaluation, including careful reasoning about confounding, leakage, and generalization.
• Advanced proficiency in Python and modern ML frameworks such as PyTorch or JAX, with the ability to build the data, training, and evaluation systems needed to pursue your own ideas.
• A strong record of first-author, peer-reviewed publications in leading machine learning, time-series, digital health, or related venues.
• Clear communication, intellectual honesty, and a collaborative, low-ego approach to technical disagreement and feedback.

Nice to have:

• Experience with longitudinal time-series, wearable sensors, physiological signals, or other real-world health data.
• Expertise in multimodal learning, generative modeling, forecasting, personalization, efficient or on-device models, or ML systems.
• Experience translating research advances into prototypes or production product capabilities.
• Familiarity with clinical research, causal inference, or evaluation using research and clinical datasets.
• Genuine interest in using machine learning to improve health rather than applying foundation models as an end in themselves.