Artificial Intelligence Researcher

About the job

Join a company that developed a virtual fitting room platform using AI, computer vision, and AR, letting online shoppers visualize clothes on their own body or diverse models.

What You'll Be Doing

You will be responsible for researching and developing novel deep learning models that generate realistic, temporally consistent video content to help customers visualize apparel in motion. You will work at the bleeding edge of technology, taking ideas from research papers to Proof of Concept (PoC) and eventually to production at massive scale.

• Lead Generative Video Research: Design and train advanced generative models for video synthesis (e.g., Video Diffusion Models, Temporal Attention layers) to create high-fidelity virtual try-on videos.
• Solve Complex Vision Problems: Tackle challenges specific to video generation, such as temporal consistency, motion preservation, and identity retention across frames.
• Innovate & Prototype: Rapidly develop future-facing Proofs of Concept (PoCs) utilizing Diffusion models, LLMs, and 3D-aware architectures.
• Scale to Production: Collaborate with engineering teams to optimize complex models for low-latency inference, ensuring they can serve millions of users on Walmart.com.
• Stay Ahead of the Curve: Continuously survey the latest academic literature in GenAI, CV, and Graphics, applying key insights to our domain.

Requirements

• M.Sc. or Ph.D. in Computer Science, Electrical Engineering, or a related field (focus on Deep Learning/Computer Vision).
• 5+ years of hands-on research experience in Deep Learning and Computer Vision.
• Mandatory GenAI Video Experience: Proven track record researching and training models for video generation. Familiarity with architectures like Video LDM, Sora-like transformers, or Motion Modules is essential.
• Strong Technical Stack: Proficiency in Python and deep learning frameworks (PyTorch or TensorFlow).
• Publication Record: Experience publishing in top-tier conferences (CVPR, ICCV, ECCV, SIGGRAPH, NeurIPS) is a strong plus.
• Experience with Human-Centric AI (Pose estimation, Human mesh recovery, SMPL).
• Knowledge of 3D Vision technologies (NeRF, 3D Gaussian Splatting).

Job No. 417557

CV to hilas@nisha.co.il