The role
From Google's own posting.
The Health Intelligence team is focused on developing frontier technologies to help everyone live healthier, happier and longer lives. We build large sensor foundation models to drive scientific discovery for novel health biomarkers. In this role, you will be working with world-scale multimodal datasets consisting of longitudinal sensor data, health agent interactions and clinical health records data.
In this role, you will focus on driving comprehensive model optimization, novel model architectures (transformers, state-space-models, etc.) and training methods, transform large-scale time-series data and experimentation systems, study training dynamics, design evaluations, and turn successful research into systems that operate reliably in production. You will develop autoresearch agents that accelerate research workflows and scientific discovery. Where existing datasets cannot answer a question, you will work with expert teams to define new endpoints, commission studies, or collect new data.
As a team we work at the forefront of technology and have the space to innovate with it. All of us are personally invested in the fitness and health space we work in and are motivated by a desire to meaningfully improve our users' lives.
The Health Platforms and Devices team builds innovative products and services that help our users live longer, healthier lives. We bring together the best of Google technologies and AI, health behavior science, and user-centered design to help users organize the health and wellness data, get insight from it, and take action toward their health goals. We do this with a suite of apps, services, and health wearables. We aim to make consumer health more personal, proactive, and actionable.
Design, train, and evaluate machine learning models, owning the full experimentation loop.
Develop automated-research agents capable of running quantitative evaluations, generating hypotheses, and executing computational experiments.
Develop evaluation frameworks that test scientific reasoning, temporal understanding, calibration, generalization, data leakage, and real-world utility.
Work with scientists to translate research questions into measurable endpoints and experimental designs, and provide technical leadership through architecture reviews and mentoring.
Provide decisive technical leadership by taking ownership in team settings, actively steering technical agendas, and making concrete decisions to overcome technical stalemates.
Minimum qualifications:
Bachelor’s degree or equivalent practical experience.
5 years of experience in machine learning research or research engineering, including experience leading technical projects.
3 years of experience training, adapting, or evaluating large-scale foundation models, and building data pipelines for heterogeneous datasets.
3 years of experience with modern machine learning frameworks (e.g., JAX, PyTorch, TensorFlow) and distributed training on accelerators.
3 years of experience designing evaluations, metrics, and controlled ablations for research projects.
Preferred qualifications:
Master's degree or PhD in Computer Science or related technical field.
Experience modeling longitudinal or multimodal real-world data (e.g., audio, wearable sensor data, health records).
Experience profiling and debugging distributed training on TPU or GPU clusters (including handling data noise and training dynamics).
Domain knowledge for health or fitness, combined with experience in scientific study design and causal inference.
Record of influential research, deployed ML systems, open-source contributions, or technical leadership in an advanced ML organization.
Bachelor’s degree or equivalent practical experience.
5 years of experience in machine learning research or research engineering, including experience leading technical projects.
3 years of experience training, adapting, or evaluating large-scale foundation models, and building data pipelines for heterogeneous datasets.
3 years of experience with modern machine learning frameworks (e.g., JAX, PyTorch, TensorFlow) and distributed training on accelerators.
3 years of experience designing evaluations, metrics, and controlled ablations for research projects.