[Skip to main content](#main-content)

[OpenNash Jobs](https://jobs.opennash.com/)
[Hiring report](/reports/hiring/) · [All Xiaomi jobs](https://jobs.opennash.com/companies/xiaomi/) · [Search jobs](/search/)

Xiaomi · 北京

# 大模型推理框架开发工程师

北京 

Listed on Xiaomi ’s official careers site. Posted August 11, 2026. Check that it is still open and apply there.
[Apply on Xiaomi ’s careers site →](https://xiaomi.jobs.f.mioffice.cn/campus/position/7670924463541782810/detail)

[OpenNash Jobs](https://jobs.opennash.com/) / [Xiaomi ](https://jobs.opennash.com/companies/xiaomi/) / 大模型推理框架开发工程师

## Job description
- 负责基于 vLLM、SGLang 等技术的大模型推理框架的开发工作，包括但不限于设计和实现高效的推理引擎架构，优化推理过程中的计算流程，提高推理性能，以满足不同规模大模型的实时推理需求；
- 复杂大模型框架性能评估和调优工作，构建并完善性能监控体系，通过实时监测训练指标，定位性能瓶颈，提出优化方案，确保在不同硬件上达到最优性能；
- 负责Cuda Kernel优化，开发高性能的GPU算子, 提升大模型推理性能。

## Qualifications
- 硕士及以上学历，计算机、软件工程、数学等相关专业，具备扎实计算机理论基础和丰富编程经验；
- 熟悉 C++ / Python，掌握至少一种深度学习框架（如 PyTorch），理解Transformer系统深度学习原理算法；
- 熟悉 GPU 和其他加速硬件的使用，有基本的性能调优经验，有FastTransformer、CUDA优化、TensorRT、Triton经验优先；
- 熟悉大模型推理框架的基本原理和流程，了解 vLLM、SGLang 等大模型推理框架的核心技术和实现优先；
- 有良好团队协作与沟通能力，有强烈的上进心和自我驱动，学习适应能力强，乐观自信，能挑战自我不断追求卓越。

## At a glance
- Posted by the employer: Aug 11, 2026
- First seen by OpenNash: Oct 19, 2026
- Last checked: Oct 19, 2026
- Listed for: 69 days
- Location: 北京[Apply on Xiaomi ’s careers site →](https://xiaomi.jobs.f.mioffice.cn/campus/position/7670924463541782810/detail)

## More jobs at Xiaomi 

### [27届招聘实习生-国际职能平台](https://jobs.opennash.com/companies/xiaomi/jobs/6d4d527313cd96474e53/)

北京

### [AI Agent 产品经理](https://jobs.opennash.com/companies/xiaomi/jobs/fc2f52865a7e2d84fbd7/)

北京

### [AI Agent 架构师 / Tech Lead](https://jobs.opennash.com/companies/xiaomi/jobs/35dcfe12c39db2a6a88a/)

北京

### [AI Agent产品经理](https://jobs.opennash.com/companies/xiaomi/jobs/1b2859d295836fa011a9/)

北京

### [AI Agent研究员实习生-小爱](https://jobs.opennash.com/companies/xiaomi/jobs/0c7ddc4025af5dc78885/)

北京

OpenNash lists jobs from employers’ own careers sites and does not handle applications.
Dates come from the employer where stated; “first seen” and “last checked” are our own observations.

OpenNash Jobs · Official listings with dates and links to the employer. [Weekly hiring report](/reports/hiring/) · [Methodology](/methodology/)