Relational Foundation AI Engineer: Graph & Transformer

NVIDIA; Santa Clara (CA); On-site; USD 184,000 - 357,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking an experienced ML researcher to advance Transformer and Graph Neural Network models that operate over diverse relational schemas and heterogeneous graphs. You will collaborate with world-class researchers and engineers to turn foundational research into production systems powering enterprise AI.

You will work across the full ML lifecycle—from architecture exploration and large-scale training to post-training optimization and high-performance inference—contributing to scalable,

Qualifications

• MS or PhD in ML, CS or equivalent
• Proficiency in Python and PyTorch
• At least 8 years of research experience in ML algorithm design
• Experience applying predictive models to real-world applications

Responsibilities

• Collaborate with researchers and engineers to advance Transformer and GNN models over relational schemas and heterogeneous graphs
• Work on high-impact use cases such as forecasting, entity matching, customer retention and fraud detection
• Contribute to scalable and adaptable ML solutions across the full lifecycle from design to inference acceleration
• Help build the next generation Relational Foundation Model
• Contribute to production systems leveraging NVIDIA infrastructure

Skills

Research experience

ML algorithm design

Collaborative

Education

MS or PhD in Machine Learning or Computer Science

Tools

PyTorch

Transformer

Graph Neural Networks

Job description

NVIDIA is seeking an experienced ML researcher to advance Transformer and Graph Neural Network models that operate over diverse relational schemas and heterogeneous graphs. You will collaborate with world-class researchers and engineers to turn foundational research into production systems powering enterprise AI.

You will work across the full ML lifecycle—from architecture exploration and large-scale training to post-training optimization and high-performance inference—contributing to scalable,