Research Scientist: Privacy-Preserving Large-Scale Training

Ellis Technologies, Inc.; San Jose (CA); Sur place; USD 156 000 - 316 800; Plein temps

Avantages offerts par ce poste

Medical insurance

Dental insurance

Vision insurance

401(k) savings plan with company match

Paid parental leave

Short-term and long-term disability coverage

Life insurance

Paid time off

Résumé du poste

Ellis Technologies, Inc. is seeking a Research Scientist specializing in privacy-preserving model training and architecture optimization in San Jose. The candidate will design and optimize large-scale training architectures for advanced generative models and lead performance optimization efforts across GPUs.

This role requires collaboration with research teams and entails significant responsibilities in developing robust, fault-tolerant systems. A strong background in large-scale deep learning and expertise in GPU optimization is critical, along with proficiency in PyTorch.

Competitive benefits include medical, dental, vision insurance, a 401(k) plan with company match, and more, ensuring employee wellbeing and satisfaction from day one.

Qualifications

• Strong background in large-scale deep learning systems and distributed training.
• Hands-on experience with GPU optimization including memory management.
• Experience training diffusion models at scale.
• Proficiency in PyTorch and modern distributed training stacks.
• Solid understanding of parallelism strategies.

Responsabilités

• Design and optimize large-scale training architectures for generative models.
• Lead GPU-centric performance optimization for large training systems.
• Develop and evolve distributed training strategies tailored to long-running jobs.
• Build fault-tolerant, self-healing training systems.
• Design mechanisms for fast failure detection and recovery.

Connaissances

Large-scale deep learning systems

GPU optimization

Distributed training

PyTorch

Parallelism strategies

Description du poste

Ellis Technologies, Inc. is seeking a Research Scientist specializing in privacy-preserving model training and architecture optimization in San Jose. The candidate will design and optimize large-scale training architectures for advanced generative models and lead performance optimization efforts across GPUs.

This role requires collaboration with research teams and entails significant responsibilities in developing robust, fault-tolerant systems. A strong background in large-scale deep learning and expertise in GPU optimization is critical, along with proficiency in PyTorch.

Competitive benefits include medical, dental, vision insurance, a 401(k) plan with company match, and more, ensuring employee wellbeing and satisfaction from day one.