About the Company
Video Rebirth is dedicated to revolutionizing the video generation landscape through cutting-edge AI technology. Our mission is to create innovative solutions that enhance video production efficiency while maintaining the highest quality standards. We foster a collaborative and inclusive culture that values creativity and continuous learning.
About the Role
The Senior Engineer, Video AI Data Pipeline will be responsible for designing, dveloping, deploying, and optimizing end-to-end data pipelines for video generation models, ensuring a stable and scalable system to support the entire model training process.
Responsibilities
- Design, develop, deploy and optimize end-to-end data pipelines for video generation models.
- Develop video data processing pipelines and optimize operator performance, covering data collection, denoising, enhancement, deduplication, OCR, multi-modal alignment, and quality evaluation to ensure efficiency and data quality.
- Implement and iterate core algorithms (OCR, optical flow estimation, motion detection, key frame extraction, etc.) to adapt to large-scale video data processing and support core pipeline functions.
Qualifications
- Bachelor’s degree or above in Computer Science, Data Science or related fields.
- 2+ years of experience in foundation model data pipelines or audio & video data processing.
Required Skills
- Proficient in Python, PyTorch, and audio & video processing tools (FFmpeg, OpenCV, PyAV, etc.).
- Experienced in large-scale (TB/PB-level) audio & video data processing.
- Familiar with multi-modal data processing and video-text alignment methods (CLIP, VideoCLIP).
- Familiar with data storage solutions (S3, OSS, HDFS, Milvus) and experienced in storage performance optimization.
- Strong self-learning ability, good coding standards and documentation habits, and effective team collaboration skills.
Preferred Skills
- Proficient in distributed processing frameworks (Spark, Ray, Dask) and workflow scheduling tools (Airflow).
- Proficient in data quality evaluation (video clarity, multi-modal alignment, etc.) and troubleshooting pipeline performance/data quality issues.
- Hands-on experience in containerized deployment (Docker, K8s).
Pay range and compensation package
Competitive salary based on experience and qualifications.
Work Location
Shenzhen / Hong Kong