KA
Listed on Krea AI’s own careers site. You apply with them directly — we never stand between you and the employer.
What this role is
This role focuses on advancing the aesthetics and quality of Krea's image generation models through large-scale posttraining techniques, including preference optimization and reinforcement learning. It suits researchers with deep expertise in diffusion model finetuning and infrastructure who want to work on cutting-edge generative AI tools with a team that values both technical excellence and creative sensibility.
Our summary, not Krea AI’s wording. The full posting is on their site.
Skills this role names
Log in to see which of these are already on your profile.
What they ask for
Required
- Proven posttraining work on diffusion models for image or video generation
- Large-scale model training, inference, and optimization experience
- Strong understanding of LLM and diffusion posttraining pipelines and algorithms
- Strong PyTorch proficiency
- Distributed training experience with FSDP, tensor parallelism, pipeline parallelism, and other parallelism strategies
- Low precision training and inference knowledge
- VLM model training experience
- Ability to identify model weaknesses and design concrete evaluations and reward functions
Nice to have
- Knowledge of fast inference engines like vLLM and SGLang
- Understanding of RL frameworks like SLIME, MILES, Tinker, and Verl
- Experience with RL infrastructure optimization techniques
- Familiarity with related fields such as representation learning and robotics research