Microsoft Principal Software Engineer-distributed training system 

China, Beijing, Beijing 

98731070

• Bachelor, Master, PhD degree in CS/EE or related areas is required.
• 6+ years of industry experiences in software engineering.
• Solid experience of shipping high performance C++, CUDA, python, C#, or equivalent language code.
• Experience with machine learning and TensorFlow/PyTorch distributed training is preferred.
• Domain knowledge of ads, search or content services is a plus.
• Quick learning and solid problem solving and debugging skills.
• Good communication skill, fluent in English (both oral and written).

• Design and implement distributed training system for trillion parameter machine learning models.
• Drive our team efforts around utilization and optimization of training and inference on GPUs.
• Design and implement streaming training and publish of trillion parameter machine learning models.
• Analyze metrics and identify opportunities based on offline and online testing, develop and deliver robust and scalable solutions.
• Collaborate with cross-functional teams to deliver high-quality solutions.