L
Large-Scale Video Understanding Research Scientist
В офисеПолная занятостьAI & Machine LearningEnterprise Software
Join a generative AI research team developing video and audio understanding pipelines for foundational video models. The role combines multimodal model research, software engineering, computer vision, and scalable data processing across petabyte-scale datasets.
Обязанности
- Fine-tune and control video-language models for video and audio understanding
- Design algorithms to balance, filter, and curate training and evaluation datasets
- Implement classic and modern algorithms for large-scale dataset processing, clustering, evaluation, and filtering
- Work with high-performance distributed systems processing petabytes of data across virtual machines
- Improve throughput, correctness, and reproducibility in scalable data pipelines
- Collaborate with researchers and product stakeholders to improve training sets and evaluation protocols using model-performance feedback
Требования
- Experience training, fine-tuning, or post-training large-scale video-language or multimodal foundation models
- Strong software engineering skills with proficiency in JAX or PyTorch
- Ability to develop and implement computer vision models for data filtering and evaluation
- Knowledge of statistics and clustering
- Experience improving the performance and maintainability of complex systems
Условия и преимущества
- Daily door-to-door shuttle services and train-station pickups
- Car-to-go subscriptions at several locations in central Israel
- Free parking
- Two on-site chef-led restaurants and an on-site bakery
- Workshops, learning opportunities, training, and access to professional platforms