Accelerate scientific discovery and engineering advances through deep neural networks for Heat-Assisted Magnetic Recording (HAMR) by using machine learning for the capture, interpolation, and optimization of highly complex optical phenomena.
Requirements
- Experience with machine learning algorithms and optical simulations (COMSOL or other equivalent software).
- Experience with Linux and Windows computing environments.
- Introductory experience with programming languages (e.g. C++, Python, Java).
- Familiarity with MPI and high-performance computing.
Responsibilities
- Gather datasets from COMSOL simulation of different plasmonic structures, pre-process them to make the input data structure feedable into the DNN model, separating proper training/validation sets.
- Research suitable DNN architecture (discriminative/generative approaches) that can be used to design plasmonic nano-antenna for HAMR application.
- Post-process, analyze, organize, and present results.
- Draw physically meaningful conclusions about configurations and expected performance and communicate findings.
- Suggest and model new configurations, materials, and modifications intended to improve the HAMR design.
Other
- Good communication and teamwork skills are needed to collaborate with multiple organizations throughout the company.
- Pursuing a Ph.D. degree in Electrical Engineering, Physics, or related field and enrolled in Fall 2026 classes.
- Travel: None
