混元机器学习平台GPU调度研发工程师(深圳/北京）
Tencent
混元机器学习平台GPU调度研发工程师(深圳/北京）
Job Description
混元机器学习平台GPU调度研发工程师(深圳/北京）
📋 Job Overview
The role involves developing Tencent's Taiji unified GPU scheduling platform for large-scale GPU clusters. Responsibilities include managing quotas, task queuing, elastic tasks, cross-cluster operations, and performance optimizations to support diverse AI workloads. The position ensures high-performance and stable operation for business scenarios like Hunyuan large models, advertising, and video accounts.
📍 Location: Shenzhen, China
🏢 Business Unit: TEG
📄 Full Description
2.优化异构AI芯片在多种任务作业之间的全局最优匹配，实现全局最佳利用率；
3.支持腾讯混元大模型、广告、视频号等众多业务场景的离线、在线GPU作业，保障平台高性能高稳定运行。
🎯 Key Responsibilities
- Participate in the development of Tencent Taiji unified GPU scheduling platform, responsible for quota management, task queuing, elastic tasks, cross-cluster functions, and performance optimization for large-scale GPU clusters with tens of thousands of cards.
- Optimize global optimal matching of heterogeneous AI chips across multiple task jobs to achieve the best overall utilization rate.
- Support offline and online GPU jobs for numerous business scenarios such as Tencent Hunyuan large models, advertising, and Video Accounts, ensuring high-performance and stable platform operation.
Locations
- Shenzhen, China
Salary
300,000 - 600,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Responsibilities
- Participate in the development of Tencent Taiji unified GPU scheduling platform, responsible for quota management, task queuing, elastic tasks, cross-cluster functions, and performance optimization for large-scale GPU clusters with tens of thousands of cards.
- Optimize global optimal matching of heterogeneous AI chips across multiple task jobs to achieve the best overall utilization rate.
- Support offline and online GPU jobs for numerous business scenarios such as Tencent Hunyuan large models, advertising, and Video Accounts, ensuring high-performance and stable platform operation.
Tags & Categories
Answer 10 quick questions to check your fit for 混元机器学习平台GPU调度研发工程师(深圳/北京） @ Tencent.
Related Books and Jobs
混元机器学习平台GPU调度研发工程师(深圳/北京）
Tencent
混元机器学习平台GPU调度研发工程师(深圳/北京）
Job Description
混元机器学习平台GPU调度研发工程师(深圳/北京）
📋 Job Overview
The role involves developing Tencent's Taiji unified GPU scheduling platform for large-scale GPU clusters. Responsibilities include managing quotas, task queuing, elastic tasks, cross-cluster operations, and performance optimizations to support diverse AI workloads. The position ensures high-performance and stable operation for business scenarios like Hunyuan large models, advertising, and video accounts.
📍 Location: Shenzhen, China
🏢 Business Unit: TEG
📄 Full Description
2.优化异构AI芯片在多种任务作业之间的全局最优匹配，实现全局最佳利用率；
3.支持腾讯混元大模型、广告、视频号等众多业务场景的离线、在线GPU作业，保障平台高性能高稳定运行。
🎯 Key Responsibilities
- Participate in the development of Tencent Taiji unified GPU scheduling platform, responsible for quota management, task queuing, elastic tasks, cross-cluster functions, and performance optimization for large-scale GPU clusters with tens of thousands of cards.
- Optimize global optimal matching of heterogeneous AI chips across multiple task jobs to achieve the best overall utilization rate.
- Support offline and online GPU jobs for numerous business scenarios such as Tencent Hunyuan large models, advertising, and Video Accounts, ensuring high-performance and stable platform operation.
Locations
- Shenzhen, China
Salary
300,000 - 600,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Responsibilities
- Participate in the development of Tencent Taiji unified GPU scheduling platform, responsible for quota management, task queuing, elastic tasks, cross-cluster functions, and performance optimization for large-scale GPU clusters with tens of thousands of cards.
- Optimize global optimal matching of heterogeneous AI chips across multiple task jobs to achieve the best overall utilization rate.
- Support offline and online GPU jobs for numerous business scenarios such as Tencent Hunyuan large models, advertising, and Video Accounts, ensuring high-performance and stable platform operation.
Tags & Categories
Answer 10 quick questions to check your fit for 混元机器学习平台GPU调度研发工程师(深圳/北京） @ Tencent.
Related Books and Jobs