Krea
ML Researcher - Image / Video Diffusion
San Francisco · On-site · Full-time · Posted Sep 1, 2026
- Experience
- Experienced researcher; proven track record working with image or video models at scale. No years specified.
- Visa sponsorship
- Open to sponsoring international visas where possible (e.g., STEM OPT, OPT, H-1B, O-1, E-3).
About the job
Trains and improves large-scale image and video diffusion models, optimizing distributed training and model quality on GPU clusters.
What you'll do
- Train image and video diffusion models and optimize large distributed training runs.
- Implement and improve distributed training strategies; debug failures and build fault-tolerance solutions.
- Improve model quality and reliability through data, architecture, training pipelines, experiments, and evaluation.
What they ask for
- Proven track record working with image or video models at scale.
- Strong proficiency in PyTorch and understanding of its inner workings.
- Strong background in distributed training strategies and experience profiling and debugging large distributed training.
- Knowledge of low-precision training/inference and diffusion training across pretraining, midtraining, preference optimization, and reinforcement learning.
Nice to have
- Publications or open-source contributions.
- Experience or interest in related fields such as LLM, VLM, representation learning, and robotics research.
Tools
PyTorch, FSDP, CP, SP, USP, TP, EP, FP8
AI focus
Research and engineering for image and video diffusion generation, including large-scale training, optimization, and evaluation.
Model-written summary · the original posting is authoritative