Summary

I'm a Senior HPC AI Engineer with expertise spanning GPU kernel optimization (CUDA, Triton: achieving 7× speedup, 5× less memory), distributed training at scale, and ML research collaboration. I develop high-performance solutions across the full stack, from low-level kernel optimization to infrastructure deployment and research support. My experience on supercomputers like Jean Zay and GB200 NVL72 covers performance optimization, infrastructure automation, and hands-on collaboration on cutting-edge AI research projects across multiple scientific domains (NLP, computer vision, astrophysics, climate modeling). I also provide training and consultancy to research institutions on AI optimization and deep learning techniques.

Work Experience

Teachings

During my experience at IDRIS, one of my task was to provide teaching courses to universities, other institutes, and even private companies. Here is an overview of teachings I did during my time at IDRIS.

• Specialization of Large Language Models: Prompt Engineering & Fine-tuning (Manager of this course)
• Optimization of Deep Learning for Supercomputers
• Other IDRIS formal courses
• Miscellaneous

and many other...

See all those who trusted me and my colleagues for our teachings

Education