ML Researcher - Image / Video Diffusion

Krea is seeking an experienced ML Researcher with engineering skills to train large-scale image and video diffusion models on GPU clusters. You will optimize distributed training runs, implement advanced training strategies, and improve model quality through experimentation and data pipeline enhancements.

• Train diffusion models for image and video generation on large GPU clusters
• Optimize and profile large distributed training runs
• Implement and improve distributed training strategies (FSDP, CP, SP, TP, EP)
• Improve model quality and reliability through data, architecture, and pipeline enhancements
• Debug distributed training errors and implement fault tolerance solutions
• Ablate architecture, attention, optimizer, data, and algorithmic choices