Senior Software Engineer, Computer Vision (Knowledge Distillation)

This role focuses on designing and implementing highly efficient AI systems for generative tasks around diffusion models and vision, with emphasis on training stability and scalable deployment.
You will advance algorithm acceleration through knowledge distillation, architecture redesign (dynamic MoE routing, sparse attention), and low-bit quantization.
Responsibilities include optimizing diffusion sampling, latent optimization, and autoregressive components for faster, cheaper inference.
You will prototype and ship hardware-efficient models and tooling, enabling large-scale multi-GPU, multi-machine deployment.
Strong collaboration with ML research, infrastructure, and product teams is essential to translate GenAI capabilities into products.
Ideal candidates bring a track record in diffusion models, model compression, and efficient architectures, plus excellent communication in fast-paced settings.