Lead ML Engineer: Physics-Informed AI for Precision Tech

WESTERN DIGITAL (SINGAPORE) PTE. LTD.; Singapore; On-site; SGD 150,000 - 230,000; Full time

Job summary

WESTERN DIGITAL (SINGAPORE) PTE. LTD. seeks an ML Technical Lead to deploy physics-informed AI in precision product development.

You will design CNN, U‑Net, and ViT systems for inspection, enable real-time anomaly detection, and own the end-to-end ML platform from experimentation to deployment. You will validate PINNs, surrogate models, and Bayesian design, lead architectural decisions, and mentor engineers while collaborating with domain scientists to translate requirements into robust

Qualifications

• Bachelor's or Master's degree in Artificial Intelligence, Machine Learning, Computer Science, or related field.
• AI major or strong AI research focus preferred.

Responsibilities

• Design and own CNN, U-Net, and ViT systems for precision inspection, measurement, and defect classification.
• Own real-time anomaly detection from sensor and time-series data streams.
• Validate and deploy PINNs and surrogate model pipelines for product development.
• Own ML platform responsibilities including MLflow, Docker, AWS EKS, CI/CD, observability, and monitoring.
• Provide technical direction, lead design reviews, and mentor junior team members.

Skills

Python

PyTorch

Computer Vision

Anomaly Detection

Surrogate Modeling

Active Learning

MLOps

Docker

AWS EKS

LangFuse/PortKey

Education

Bachelor's or Master's in AI/ML/CS

Tools

MLflow

Docker

AWS EKS

LangFuse

PortKey

Job description

WESTERN DIGITAL (SINGAPORE) PTE. LTD. seeks an ML Technical Lead to deploy physics-informed AI in precision product development.

You will design CNN, U‑Net, and ViT systems for inspection, enable real-time anomaly detection, and own the end-to-end ML platform from experimentation to deployment. You will validate PINNs, surrogate models, and Bayesian design, lead architectural decisions, and mentor engineers while collaborating with domain scientists to translate requirements into robust