AI Research Scientist, Scientific ML
Core
Design and deploy physics-constrained neural network methodologies (PINNs, Bayesian design, causal ML) to drive real laboratory experiments and product development for storage systems.
Role type
Senior IC scientific ML researcher (physics-informed)
Builds
Physics-constrained loss function architectures, digital twin ML components, validated research prototypes, and acquisition functions for autonomous experiment selection.
Domain
Scientific machine learning, physics-informed neural networks (PINNs), Bayesian experimental design, causal inference, materials science/semiconductor domain.
Deliverable
production ML models (via careerplan.io/jobs/744000153657449-ai-research-scientist-scientific-ml-at-western-digital)
Required skills
PINNs methodology design, physics-constrained loss function architecture, Bayesian deep learning, active learning acquisition function design, structural causal models (SCM), causal discovery, PyTorch or JAX expert implementation, prototype-to-documentation handoff
Preferred skills
Diffusion models for synthetic data generation, graph neural networks (GNN), neural ODEs, foundation model fine-tuning, RL for scientific discovery
Technologies
PyTorch, JAX, diffusion models, VAEs, GNN, Neural ODEs
Responsibilities
Design physics-constrained loss function architectures; validate digital twin ML components against domain physics; lead research into Bayesian experimental design and active learning; develop causal inference frameworks for reliability root cause analysis; produce validated research prototypes with complete technical documentation.
Seniority
Senior, hands-on IC
