Research Engineer — Generative Models (Remote — Worldwide)

Stability AI is seeking a Research Engineer to push the boundaries of open-source generative AI across modalities including image, video, and audio. You will join a mission-driven team dedicated to building the next generation of Stable Diffusion and other foundation models, ensuring high-performance AI is accessible to everyone. This role focuses on scaling model training and optimizing inference to define the future of creative technology.

- Develop and implement novel architectures for text-to-image, text-to-video, and multi-modal foundation models. - Optimize large-scale training runs on thousands of GPUs, focusing on stability, throughput, and memory efficiency. - Conduct rigorous evaluations and benchmarks to measure model performance, safety, and alignment. - Collaborate with the infrastructure team to improve internal tooling for model training and deployment. - Translate cutting-edge research papers into production-ready code for the open-source community. - Contribute to the release of open-weight models, documentation, and technical reports for global developers.

- Master’s or PhD in Computer Science, Mathematics, or a related field with a focus on Deep Learning. - Proven track record of training large-scale generative models (GANs, Diffusion, or Autoregressive Transformers). - Expert level proficiency in PyTorch and experience with distributed training frameworks like DeepSpeed or FSDP. - Strong understanding of linear algebra, calculus, and probability theory as applied to generative modeling. - Experience managing large-scale datasets and implementing efficient data pipelines for multi-node GPU clusters. - Contributions to open-source AI projects or a strong portfolio of peer-reviewed publications (NeurIPS, ICML, CVPR). - Ability to work asynchronously across global time zones in a fast-paced, remote-first environment.

Equity, compute budget, conference travel.