Applied SciML (Scientific Machine Learning ) Engineer

Govserviceshub; Santa Clara (CA); On-site; USD 100,000 - 140,000; Full time

Job summary

A leading technology firm in Santa Clara is seeking an Applied SciML Engineer with expertise in scientific machine learning and computational geometry. The ideal candidate will have a minimum of 3 years of deep learning experience and proficiency in PyTorch. This role focuses on building production ML systems for physical simulation data. Candidates should have a strong background in model development, optimization, and application to scientific datasets.

Qualifications

• Minimum 3 years of experience in deep learning.
• Experience with CAD geometries or CFD simulations.
• Proficiency in PyTorch and related frameworks.

Responsibilities

• Build production ML systems for physical simulation data.
• Optimize multi-GPU training for efficiency.

Skills

Core Deep Learning Experience

Hands-on deep learning training with scientific datasets

End-to-end model development

Multi-GPU training and GPU memory optimization

PyTorch proficiency

Computational Geometry knowledge

Job description

Applied SciML (Scientific Machine Learning) Engineer

Santa Clara, United States | Posted on 10/16/2025

Role: Applied SciML (Scientific Machine Learning) Engineer
Location: Santa Clara, CA

MUST HAVE:

Job Description

• We're looking for an engineer with deep expertise in scientific machine learning and computational geometry to build production ML systems for physical simulation data.

Requirements

• Core Deep Learning Experience (3+ years)
• Hands‑on deep learning training with scientific datasets: CAD geometries, CFD simulations, or similar physical data
• End‑to‑end model development: data preparation, training, hyper‑parameter tuning
• Multi‑GPU training and GPU memory optimization
• PyTorch and PyTorch Lightning proficiency

Technical Depth

Demonstrated experience (via publications in top venues or 3+ years industry experience) in one or both areas:

• Computational Geometry: PointNet, DGCNN, TripNet, MeshGPT, or similar geometric deep learning architectures
• Scientific ML: PINNs, DeepONet, FNO, Transformers for physics, SCoT, Poseidon, or related physics‑informed models
• Evidence of impact — publications in leading ML/computational science conferences/journals or a proven track record building SciML systems in industry