# 混元大模型推理研发高级工程师（深圳/北京/上海/杭州）

External opening: you cannot apply through AI Native Match; apply on the official site.

Apply through the official channel. The full external receipt stays in your workspace. When signed in, three minimal event types explicitly recorded through the CLI are reported by default: selection, self-reported application and channel observation. Browsing or saving a role does not report an event; external applications do not appear under My applications.

- Company: 腾讯 (Stage unknown)
- Locations: 北京
- Role: Engineering
- Apply on the official site: https://careers.tencent.com/jobdesc.html?postId=1958076178912923648

## Role description from source

1.配合算法工程师，推动深度学习相关算法的落地，打造高吞吐、低延时的推理系统；
2.优化大模型推理性能，提升吞吐并控制成本；
3.优化大模型推理框架，提升框架易用性和可调试性。
职位要求
1.熟练掌握 C/C++、Python语言，有计算机体系结构背景或软件开发背景，熟悉系统性能调优的方式；
2.具备基础的GPU编程能力，包括但不限于Cuda、OpenCL，熟悉至少一种GPU加速库，如cublas、cudnn、cutlass等；
3.具备Tensorrt/FasterTransformer/Tensorrt-llm/vllm/sglang等推理引擎的使用和性能优化经验；
4.熟悉各类深度学习网络和算子底层实现细节，具备训推模型调试、调优实操经验优先；
5.熟悉CPU/GPU异构加速瓶颈分析方法，有服务器端 AI 芯片、GPU加速经验优先；
6.熟悉分布式推理常用加速方法，有超大模型分布式部署经验优先。
工作年限（官方顶部字段）
一年以上工作经验

---
opening_id: fad56ccd-205b-445d-ad3a-3ed0535bda0e

`anm openings save --id fad56ccd-205b-445d-ad3a-3ed0535bda0e`