Research Engineer – World Modelling

About the role

MBZUAI is seeking a Research Engineer to design and maintain scalable video data pipelines, develop robust preprocessing and annotation workflows for world‑model training, and optimise distributed systems that support multimodal foundation models. The role focuses on improving inference efficiency for real‑time interaction and translating research prototypes into production‑ready, reproducible systems.

Key responsibilities

• Design, implement and maintain scalable video data pipelines for large‑scale training.
• Develop data preprocessing, transformation and synthesis workflows for world‑model training.
• Build high‑quality data annotation pipelines to ensure accurate labeling across massive datasets.
• Support training of multimodal foundation models by developing and optimising distributed training systems.
• Improve inference and serving efficiency for real‑time interaction through model optimisation and system tuning.
• Monitor system health, debug performance issues and contribute to large‑scale optimisation.
• Collaborate closely with research teams to turn experimental ideas into reliable infrastructure.
• Integrate novel research prototypes into production‑ready systems and ensure reproducibility at scale.

Required profile

• MSc or PhD in Machine Learning, Computer Science, or equivalent industry experience.
• Proven experience in large‑scale data collection, cleaning and transformation for multimodal datasets (video, audio, text).
• Hands‑on experience with web‑scraping frameworks such as Scrapy, Selenium, Playwright or BeautifulSoup.
• Experience training large models (LLMs or diffusion models) on high‑performance clusters.
• Practical knowledge of state‑of‑the‑art video generative models (e.g., Sora, Veo2, MovieGen, CogVideoX).
• Strong systems engineering expertise in deep‑learning frameworks, especially PyTorch.
• Demonstrated ability to solve complex system‑level challenges and debug issues across the training/inference stack.

Required skills

• PyTorch
• Scrapy
• Selenium
• Playwright
• BeautifulSoup
• Distributed training systems (e.g., PyTorch Distributed)
• Video diffusion models (Sora, Veo2, MovieGen, CogVideoX)