Training Performance Engineer for ML & CUDA Optimizations

Jaide Health; Paris; Hybride; EUR 60 000 - 80 000; Plein temps

Avantages offerts par ce poste

Inclusive workplace culture

Weekly lunch stipend

Full health and dental benefits

100% parental leave top-up

Personal enrichment benefits

Remote-flexible work options

6 weeks of vacation

Résumé du poste

Jaide Health is seeking a Performance Engineer to optimize the performance of advanced language models within a vibrant team dedicated to AI innovation. You'll focus on enhancing training throughput and eliminating bottlenecks through cutting-edge practices.

Your responsibilities include designing scalable software, writing low-level kernels, and researching data infrastructure. The ideal candidate should have strong software engineering skills and experience with Python. Enjoy a hybrid work environment with generous perks including six weeks of vacation!

Qualifications

• Extremely strong software engineering skills.
• Proficiency in Python and ML frameworks such as JAX and PyTorch.
• Experience writing kernels for GPUs using CUDA and Triton.

Responsabilités

• Optimize performance of advanced language models.
• Develop training and profiling tools for model performance.
• Research and experiment on supercompute and data infrastructure.

Connaissances

Software engineering

Python

CUDA

Machine Learning frameworks (JAX, PyTorch)

Distributed training strategies

Transformers

Description du poste

Jaide Health is seeking a Performance Engineer to optimize the performance of advanced language models within a vibrant team dedicated to AI innovation. You'll focus on enhancing training throughput and eliminating bottlenecks through cutting-edge practices.

Your responsibilities include designing scalable software, writing low-level kernels, and researching data infrastructure. The ideal candidate should have strong software engineering skills and experience with Python. Enjoy a hybrid work environment with generous perks including six weeks of vacation!