📋 Job Overview

The role focuses on developing inference deployment solutions for AI large models like DeepSeek on Kubernetes, integrating deeply with customer scenarios. It involves implementing AI infrastructure capabilities in TKE, including designing and developing AI-related workloads to reduce user costs. Additionally, the position emphasizes optimizing compute, network, and storage resources to enhance training and inference efficiency, while addressing inference stability, scheduling, framework optimization, and GPU pooling to lower costs and boost performance.