About Black Forest Labs

• We’re the team behind Latent Diffusion, Stable Diffusion, and FLUX—foundational technologies that changed how the world creates images and video.
• We’re creating the generative models that power how people make images and video—tools used by millions of creators, developers, and businesses worldwide.

Why This Role

• You'll train large-scale diffusion models for image and video generation, exploring new approaches while maintaining rigor.

What You’ll Work On

• Trains large-scale diffusion transformer models for image and video data, working at an impactful scale.
• Rigorously ablates design choices to produce actionable insights.
• Reasons about speed-quality tradeoffs of neural network architectures.
• Fine-tunes diffusion models for specialized applications like upscalers and inpainting models.

What We’re Looking For

• Hands-on experience training large-scale diffusion models for image and video data.
• Experience fine-tuning diffusion models for specialized applications.
• Deep understanding of evaluating generative models and proficient in PyTorch.
• Strong understanding of distributed training techniques.

• Have experience writing forward and backward Triton kernels.
• Proficiency with profiling, debugging, and optimizing GPU operations.
• Know the performance characteristics of different architectural choices at scale.
• Published research contributing to generative models.

How We Work Together

• Obsessed. Frontier research lab with a focus on science, understanding, and product quality.
• Low Ego. The best idea wins, credit shared, no task too small.
• Bold. Taking ambitious bets and shipping without waiting for perfection.
• Kind. Treating each other with warmth and empathy.