Software Engineer - video diffusion models

Software Engineer / Research Scientist – Video Diffusion Models & Character Motion Control

Our client is revolutionizing video generation by solving one of the toughest problems in the space: controlling how characters move. While most AI video models produce awkward, unpredictable motion, they're changing the game with their own motion control tech — a system that lets creators generate consistent, guided character movement by combining text prompts with real-world acting videos.

We’re searching for a Software Engineer or Scientist with deep expertise in video diffusion models, generative AI, and motion control to join their cutting-edge team.

The role (to be further defined but a rough guide)

• Develop and improve video diffusion architectures to generate realistic, human-like motion
• Integrate text and video inputs to ensure controlled and coherent character behavior
• Work on pipelines that translate acting videos into motion-guiding data
• Collaborate with research and engineering teams to push the boundaries of video-based generative AI

Experience required

• Strong background in video diffusion models, human motion synthesis, or video generation
• Experience with multimodal input fusion — text + video
• Proficiency in deep learning frameworks (e.g., PyTorch, TensorFlow) a nice to have
• Research or applied experience in generative video, motion transfer, or pose estimation
• Passion for building tools that bring creativity and control to AI video generation

For more information, send your CV to this advert or email it to a.kanji@nonstopconsulting.com