oddin - Applied Science Intern - World Model
Requirements
• Pursuing PhD! (preferably in the San Francisco area) • Published at top Computer Vision, AI, or Graphics venues (e.g., CVPR, ICML, ICCV, Siggraph, NeurIPS). • Demonstrated hands-on experience with building and running generative CV models (e.g., GANs, DiT, VAE). • Solid understanding of neural architectures and paradigms (e.g., Transformers, Denoising Diffusion Models, RNNs, Sequence Models, CNNs). • Solid understanding of VAEs (e.g., ELBO). • Basic understanding of Reinforcement Learning. • Proficiency in Python and PyTorch. • This role offers a unique opportunity to shape the future of interactive video content, where digital humans can engage in meaningful and dynamic interactions with users. If you're a passionate ML expert with a drive to innovate and create immersive experiences, we encourage you to apply.
Responsibilities
• Explore how to use World Models for understanding, simulations, and ultimately generation of sport or eSport matches (e.g., soccer, DOTA). • Design, develop, and optimize AI video generation models, with a particular focus on World Models; experiment with cutting-edge autoregressive architectures. • Develop and implement state-of-the-art algorithms for synthesizing sport matches. • Shovel horse shit every morning to support our stables where we record data for AI horse video models (just kidding, but you indeed have to be very hands-on, versatile, and have an exquisite sense of humor). • Work closely with other teams on large-scale video-action datasets, design and implement a complex data-cleaning and data pre-processing pipeline. • Define robust validation strategies and implement custom evaluation metrics comparing synthetic vs. real gameplay. • Stay on the bleeding edge of the relevant literature, e.g., CVPR, NeurIPS, ICML, ICCV, and help to align it with our roadmap.
Apply in one click
Upload My Resume
Drop here or click to browse · Tap to choose · PDF, DOCX, DOC, RTF, TXT