机器学习平台存储研发工程师-Data AML
Core
Design and develop storage components for a machine learning system serving NLP, CV, and Speech model training, evaluation, and inference.
Role type
Senior IC machine learning platform storage engineer
Builds
Multi-level storage systems integrating GPU memory, local memory, distributed memory/disk, and remote storage (HDFS/Object Storage) for efficient data flow.
Domain
Cloud-native infrastructure + Machine Learning
Deliverable
production ML models
Required skills
C++, Go, Python, Linux, Kubernetes, PV/CSI, Distributed systems design
Preferred skills
vLLM, SGLang, PyTorch, Large model inference/training, Distributed caching (Alluxio, JuiceFS), NVLink, RDMA, NC
Responsibilities
Design and implement multi-level storage systems; Manage and monitor storage in Kubernetes clusters; Build and manage multi-region/multi-cloud disaster recovery systems.
Seniority
Senior, hands-on IC