端侧大模型推理与量化资深工程师
北京
全职
通用智能板块
职位描述
1、负责大模型与语音算法的量化、剪枝、蒸馏等模型精简与加速技术，完成在高通等移动端的转换、部署，保障端侧低延迟与高精度； 2、设计并实现端侧推理引擎，负责关键推理优化策略（如 Prompt Cache、投机解码、Flash Attention 等）的工程落地，结合图优化与算子适配，最大化 NPU/CPU/DSP硬件利用率； 3、负责跨平台推理 SDK 的架构设计、封装与迭代，提供稳定、高效的端侧推理接口； 4、负责推理链路延迟、精度与功耗问题的定位，并给出优化方案，推动端侧推理性能指标达成； 5、开发并迭代模型转换、精度与性能自动化测试工具链，提升算法到端侧的交付效率。
职位要求
1、硕士及以上学历，人工智能、计算机、电子、数学等相关专业； 2、熟练掌握深度学习与大模型压缩和加速技术，包括量化、剪枝、蒸馏等，有端侧部署与精度调优实践经验； 3、熟练掌握大模型推理优化策略（如 Prompt Cache、投机解码、Flash Attention 等），熟悉 SNPE/QNN 或者MTK 等主流平台推理框架； 4、熟悉移动端与 Native 开发，熟练使用现代 C++，熟悉 Android Native / HAL 层开发调试与系统级性能分析； 5、具备良好的跨团队协作与推动能力，能够与算法团队、芯片供应方紧密沟通，推动模型部署的端到端性能和效果优化； 6、熟悉 ARM NEON / DSP intrinsic 等端侧底层加速技术者优先； 7、熟悉 ONNX模型，熟悉 MNN/NCNN 等通用推理框架及其原理者优先； 8、熟悉 VAD、唤醒、ASR 等语音前端算法逻辑者优先。
投递
XPeng is a leading Chinese Smart EV company that designs, develops, manufactures, and markets Smart EVs that appeal to the large and growing base of technology-savvy middle-class consumers. Its mission is to drive Smart EV transformation with technology and data, shaping the mobility experience of the future. In order to optimize its customers’ mobility experience, XPeng develops in-house its full-stack advanced driver-assistance system technology and in-car intelligent operating system, as well as core vehicle systems including powertrain and the electrical/electronic architecture. XPeng is headquartered in Guangzhou, China. In 2021, the Company established its European headquarters in Amsterdam, along with other dedicated offices in Copenhagen, Munich, Oslo, and Stockholm.The Company’s Smart EVs are mainly manufactured at its plant in Zhaoqing and Guangzhou，Guangdong province.
For more information, please visit https://heyxpeng.com.
