# ML Researcher - Image / Video Diffusion

Hiring organization: [Krea AI](https://career.thegoodapps.co/organizations/krea-ai)

Canonical page: https://career.thegoodapps.co/jobs/bc8cf443-5ebe-4c27-a66f-2a4084276769

Listed on Krea AI's own careers site. Applications go to them directly.

- Employment type: full time
- Seniority: Mid
- Location: San Francisco, CA

## Summary

Krea is seeking a researcher to optimize and train large-scale diffusion models for image and video generation on distributed GPU clusters. This role suits experienced machine learning engineers comfortable with complex distributed training infrastructure, profiling large experiments, and collaborating closely with a creative-focused AI team.

_Our summary, not Krea AI's wording._

## Skills named

PyTorch

## Required

- Proven track record training image or video models at scale
- Strong proficiency in PyTorch and understanding of its internals
- Deep knowledge of distributed training strategies including FSDP, CP, SP, TP, and EP
- Experience profiling and debugging large distributed training runs
- Knowledge of low precision training in FP8, NVFP4, and MXFP8
- Solid understanding of diffusion model training across pretraining, midtraining, preference optimization, and RL
- Comfortable working with underspecified research goals and breaking them into concrete plans
- Ability to iterate rapidly and propose creative research directions

## Nice to have

- Publications or open-source contributions working with image or video models at scale
- Familiarity with developments in LLM, VLM, representation learning, and robotics research

Apply on Krea AI's site: https://jobs.ashbyhq.com/krea/845dce21-e270-49a2-9229-5becd256b829/application