A premier, high-growth Israeli Generative AI & Deep-Tech startup engineering cutting-edge Multimodal Video Generation platforms in strategic partnership with global tech giants, including Microsoft.
The company's breakthrough technology enables real-time generation of hyper-realistic Digital Presenters and interactive video experiences from text at massive scale, serving global enterprise organizations.
Backed by significant venture capital funding from tier-1 global funds, the company combines international operations with an elite core R&D hub in Israel, providing an advanced research environment tackling fundamental challenges across Video Diffusion, real-time inference, and generative foundation models.
Located in central Tel Aviv (walking distance from the train station), operating on a hybrid work model.
Role Description-
- Serving as a Senior Generative AI & Video Diffusion Research Scientist, joining the core AI Research group responsible for designing and training next-generation Multimodal Video Diffusion Models and real-time generative architectures.
- Leading end-to-end research lifecycles: from foundational exploration and large-scale model training down to rigorous evaluation, production optimization, and deployment.
- Solving complex generative video challenges, focusing on Visual Quality, Temporal Consistency, Identity Preservation, and Photorealism.
- Scaling large-scale distributed training runs and engineering high-efficiency, low-latency Real-Time Inference pipelines.
- Developing algorithms and architectures at the intersection of Computer Vision, Image Processing, and Deep Learning.
- Publishing, prototyping, and integrating state-of-the-art research breakthroughs into commercial enterprise platforms.
Requirements-
- Academic background meeting one of the following tracks:
- Ph.D. directly from an accredited university in Computer Science, Electrical Engineering, Deep Learning, or related field
- M.Sc. (with Thesis) from a recognized university + proven industry research experience
- B.Sc. from a recognized university + extensive senior industry research experience – Mandatory
- Proven research background in Generative Modeling, Computer Vision, or Deep Learning – Mandatory
- Hands-on practical experience in training Video Diffusion Models (e.g., Latent Diffusion, DiT, Video DiT) – Mandatory
- Deep theoretical and practical mastery of Modern Generative Architectures & Training Methodologies – Mandatory
- Strong Software Design & Programming skills (Python, PyTorch, distributed training frameworks like DeepSpeed/FSDP/Megatron) – Mandatory
- Hands-on experience with Large-Scale GPU Cluster Training and Multi-Node Infrastructure – Significant Advantage
- Proven track record in Real-Time Model Inference Optimization (e.g., TensorRT, ONNX, Model Quantization, Distillation) – Significant Advantage
- Top-tier academic publications (CVPR, ICCV, ECCV, NeurIPS, ICML, SIGGRAPH) – Significant Advantage
155085