A
ML Research Engineer - Physics - Informed AI
AppleHerzliya, Tel Aviv District, Israel
Apply Apple's ML-Physics team in Israel is looking for an ML Engineer to join a research-oriented group developing physics-informed neural networks for advanced sensing and optical measurement systems. The role focuses on training, validating, and improving neural network models that combine data-driven learning with physical modeling, simulation, and experimental measurements.
Apple designs consumer electronics that have touched millions and changed the way people interact with electronic devices around the world. The ML-Physics team works at the intersection of machine learning and physics, building models that leverage both data-driven learning and deep physical understanding to solve complex sensing and optical measurement challenges.
- Develop, train, and evaluate neural network models for physics-based sensing and reconstruction tasks
- Incorporate physical knowledge into ML models through data generation, model architecture, loss functions, constraints, or regularization
- Work with simulated and experimental datasets, including data preprocessing, augmentation, validation, and error analysis
- Build robust training pipelines using Python and modern ML frameworks such as PyTorch
- Compare model predictions against ground-truth data from simulations, optical experiments, or measurement systems
- Collaborate with physicists and optical engineers to understand underlying physical models and translate them into ML workflows
- Analyze model performance, identify failure modes, and improve generalization from simulation to real-world data
- Design experiments to evaluate model robustness, sensitivity, uncertainty, and accuracy
- MSc or PhD in Computer Science, Electrical Engineering, Physics, Applied Mathematics, or a related field
- Hands-on experience with machine learning and deep learning
- Proficiency in Python and PyTorch or a similar deep learning framework
- Strong communication skills and proficiency in English
- Ability to work well in a collaborative, highly cross-functional, and fast-paced environment
- Solid understanding of physics and the ability to apply physical principles to machine learning problems
- Strong understanding of optimization, model training, validation, and debugging of neural networks
- Experience with scientific computing libraries such as NumPy, SciPy, pandas, or similar tools