3 days ago
Responsibilities
- Design, train, and evaluate machine learning models while owning the full experimentation loop.
- Develop automated research agents that generate hypotheses, run quantitative evaluations, and execute computational experiments.
- Build evaluation frameworks for scientific reasoning, temporal understanding, calibration, generalization, data leakage, and real-world utility.
- Develop model architectures, optimization methods, training methods, and experimentation systems for large-scale time-series and multimodal data.
- Collaborate with scientists to translate research questions into measurable endpoints and experimental designs.
- Turn successful research into reliable production systems and provide technical leadership through architecture reviews, mentoring, and technical decision-making.
Requirements
- Bachelor’s degree or equivalent practical experience.
- At least 5 years of experience in machine learning research or research engineering, including leading technical projects.
- At least 3 years of experience training, adapting, or evaluating large-scale foundation models and building data pipelines for heterogeneous datasets.
- At least 3 years of experience with JAX, PyTorch, or TensorFlow and distributed training on accelerators.
- At least 3 years of experience designing evaluations, metrics, and controlled ablations for research projects.
- Preferred qualifications include a master’s degree or PhD in Computer Science or a related technical field.
- Experience with longitudinal or multimodal real-world data such as audio, wearable sensor data, or health records is preferred.
- Experience profiling and debugging distributed training on TPU or GPU clusters, including data noise and training dynamics, is preferred.
- Health or fitness domain knowledge, scientific study design, causal inference, influential research, deployed ML systems, open-source contributions, or technical leadership in an advanced ML organization are preferred.
Tech Stack
PyTorchTensorFlow
Categories
AI ResearchML Engineering
About Google
A problem isn't truly solved until it's solved for all. Googlers build products that help create opportunities for everyone, whether down the street or across the globe. Bring your insight, imagination and a healthy disregard for the impossible. Bring everything that makes you unique. Together, we can build for everyone. Check out our career opportunities at goo.gle/3DLEokh