Machine Learning Engineer

This is a fully remote position, open to applicants in California, +2 more states.

📋 Description

• Develop physics-informed surrogate models on Azure Machine Learning to forecast engineering simulation results based on design parameters.

• Rapidly evaluate candidate designs in milliseconds, ensuring that only the most promising options undergo comprehensive high-fidelity simulation.

• Create and train surrogate models, such as neural networks, Gaussian processes, gradient-boosted trees, GNNs, and PINNs, utilizing Azure GPU computing resources.

• Integrate physics-informed constraints to ensure predictions remain physically accurate.

• Establish model uncertainty and confidence scoring to identify which designs require full simulation validation.

• Update models as new simulation data becomes available.

• Deploy and manage model versions via Azure ML endpoints and the model registry.

• Continuously monitor models for drift over time.

• Evaluate the speedup of surrogate models versus full simulations to inform platform-level performance optimization.

• Collaborate with data scientists, LLM engineers, and MLOps teams to ensure dependable and efficient GPU-intensive training and simulation operations.

⛳️ Requirements

• Significant hands-on experience in building, training, and deploying machine learning models in a production environment—not merely utilizing pretrained APIs.

• Over 10 years of experience in developing machine learning solutions for physical or engineering systems, including surrogate modeling, physics-informed ML, or scientific ML.

• Proficient in Python with experience in PyTorch or TensorFlow.

• Solid understanding of relevant engineering and physics principles, as well as simulation data formats pertinent to your field.

• Experience with Azure Machine Learning or a comparable cloud machine learning platform.

• Knowledge of uncertainty quantification techniques, including Bayesian methods and ensembling.

• Direct experience with industry-standard electromagnetic (EM) or physics simulation tools is preferred.

• Familiarity with geometric deep learning, encompassing graph neural networks and mesh-based models for CAD data is advantageous.

• A background in RF/high-speed electronics or interconnect design is considered a plus.

🏝️ Benefits

• Variable pay, provided as a monetary bonus or in an alternative form.

• Medical insurance.

• Dental insurance.

• Vision insurance.

• Flexible spending accounts.

• Health savings accounts.

• Life insurance.

• Accidental death and dismemberment (ADD) insurance.

• Disability insurance.

• Retirement benefits.

• Paid vacation and time off.

• Educational assistance.

• Infertility assistance (may apply).

• Paid parental leave (may apply).

• Adoption assistance (may apply).