顶尖应届-算法优化工程师-自动驾驶
北京
校招
正式
软件研发类
职位描述
1、主导自动驾驶算法的有损优化研发，核心负责低精度训练、网络结构优化、MTP投机采样等技术的设计、实现与迭代，在可控精度损失范围内，降低模型算力消耗、存储占用，提升推理与训练效率，适配车规级部署需求； 2、深入分析自动驾驶算法的性能瓶颈，结合车规级硬件特性，设计针对性的有损优化方案，协同量化、MTP等技术，实现模型性能与硬件算力的最优匹配； 3、跟踪行业前沿优化技术，开展技术预研与落地验证，输出技术调研报告与优化方案，推动团队技术能力迭代；结合自动驾驶场景特点，优化方案适配复杂真实路况； 4、搭建优化效果评估体系，量化分析模型优化后的精度、速度、算力消耗等指标，建立精度损失阈值标准，推动优化方案的测试、迭代与上车落地，协同测试、工程、硬件团队完成闭环验证； 5、跨团队协作，对接感知、规控、硬件部署等团队，推动优化方案的协同落地。
职位要求
1、基础功底扎实：精通深度学习核心理论，深入理解神经网络训练原理、损失函数设计，熟练掌握PyTorch等主流深度学习框架，至少一种LLM后训练实践（SFT或RLHF/RLAIF/奖励建模/DPO等），重视训练-评测落地能力，能独立完成模型训练、调优全流程； 2、编程与工程能力：精通Python、C/C++编程，具备良好的代码风格与工程实践能力，熟悉Linux开发环境，能独立开发、调试优化工具与算法； 3、数学基础：具备扎实的线性代数、概率论、微积分、凸优化基础，能理解并推导优化算法核心公式，具备独立分析和解决技术难题的能力。
职位信息
课题名称及内容：
【课题名称】 硬件感知模型优化（Hardware-aware Model Optimization） 【课题内容】 核心聚焦自动驾驶算法（感知、预测、规划/端到端等）的有损优化与全流程训练，主导低精度训练、模型量化、MTP（投机采样）等核心工作，兼顾精度与性能平衡，推动优化方案上车落地。
投递
Xiaomi Corporation was founded in April 2010 and listed on the Main Board of the Hong Kong Stock Exchange on July 9, 2018 (1810.HK). Xiaomi is a consumer electronics and smart manufacturing company with smartphones and smart hardware connected by an IoT platform at its core.
Embracing our vision of “Make friends with users and be the coolest company in the users’ hearts”, Xiaomi continuously pursues innovations, high-quality user experience and operational efficiency. The company relentlessly builds amazing products with honest prices to let everyone in the world enjoy a better life through innovative technology.
Xiaomi is one of the world's leading smartphone companies. The company has also established the world’s leading consumer AIoT (AI+IoT) platform,reached 558 million smart devices connected to its platform (excluding smartphones,laptops and tablets) as of September 30 2022. Xiaomi products are present in more than 100 countries and regions around the world. In August 2022, Xiaomi was included in the Fortune Global 500 list for the fourth year in a row, ranking 266th. The company is the fastest-rising Chinese technology conglomerate during the four-year period.
Xiaomi is a constituent of the Hang Seng Index, Hang Seng China Enterprises Index, Hang Seng TECH Index and Hang Seng China 50 Index.
