AI Data Center PSU Software Engineer

该职位来源于猎聘 Join our Innovative Team

Want to do more than just imagine the ways our world will move tomorrow? Join the technology company that's transforming the future of mobility today.

About Aptiv Aptiv is a global technology company that develops safer, greener and more connected solutions enabling the future of mobility. Headquartered in Dublin, Aptiv has approximately 150,000 employees and operates 14 technical centers, as well as manufacturing sites and customer support centers in 45 countries. Visit *************

Responsibilities and duties

• Responsible for the development of embedded software and upper-level management software for AI data center PSU (Power

Supply Unit), and interfacing with the whole rack power supply and data center monitoring platform to achieve collaborative management of the entire battery/capacitor group.

负责 AI 数据中心PSU（服务器电源单元）嵌入式软件开发和上层管控软件开发，对接整机柜电源、机房监控平台，实现服务器电源单元协同管控。

• Responsible for PSU-level policy development: complete PSU state machine and mode switching, input/output telemetry, multi-PSU redundancy, current sharing and derating strategies, hot-swap management, fault management and recovery strategies.

负责PSU级策略开发：PSU完整状态机及模式切换，输入输出遥测，多PSU冗余、均流和降额策略，热插拔管理，故障管理与恢复策略。

• Responsible for the development and debugging of PSU underlying drivers and application layer logic, implementing functions such as fault diagnosis, anomaly reporting, power outage protection, and timing protection.

负责PSU底层驱动、应用层逻辑开发与调试，实现故障诊断、异常上报、断电续保、时序保护功能。

• Responsible for developing communication protocols for PSU with the host computer in the data center, PMU, and remote operation and maintenance platform, including PMBus, CAN, RS485, and MODBUS, to enable PSU data uploading, remote parameter configuration, and fault alarms.

负责 PSU与机房上位机、PMU、远程运维平台的PMBus、CAN、RS485、MODBUS通信协议开发，实现PSU数据上传、远程参数配置、故障告警。

• Responsible for developing PSU-level fault logic, interlocking protection, fault tolerance mechanisms, and thermal management linkage strategies to ensure uninterrupted power supply to the computing power cabinet 24/7.

负责PSU级故障逻辑、连锁保护、容错机制、热管理联动策略开发，保障算力机柜 7×24h 不间断供电。

• Leading software architecture design, modular development, and version iteration; completing code standardization, version management, unit testing, and system integration testing; and outputting design documents, interface documents, and iteration reports.

主导软件架构设计、模块化开发、版本迭代，完成代码规范化、版本管理、单元测试、系统联调，输出设计文档、接口文档、迭代报告。

• In conjunction with rack integration and data center deployment testing, we address software compatibility and stability issues during mass delivery, and continuously optimize battery pack efficiency and reliability.

配合整机柜集成、机房落地测试，解决批量交付中的软件兼容性、稳定性问题，持续优化PSU能效与可靠性。

• Participate in product requirement review, solution design, unit testing, and system integration debugging; output software development documents and test reports; and cooperate with hardware to complete high and low temperature, high pressure, and aging reliability tests.

参与产品需求评审、方案设计、单元测试、整机联调，输出软件开发文档、测试报告，配合硬件完成高低温、高压、老化可靠性测试。

• Follow up on industry standards and functional safety requirements for data center energy storage, optimize software stability, real-time performance, and fault tolerance, and solve software issues in mass production and field delivery.

跟进数据中心储能行业规范、功能安全要求，优化软件稳定性、实时性、容错性，解决量产及现场交付的软件问题。

Qualifications/ Requirements：

• Bachelor's degree or above in Electronic Information, Automation, Computer Science, Electrical Engineering or related fields; at least 3 years of experience in embedded application software development; experience in data center PSU projects is preferred.

本科及以上学历，电子信息、自动化、计算机、电气工程等相关专业，3 年以上 嵌入式应用软件开发经验，有数据中心 PSU项目经验优先。

• Proficient in C language embedded development, with the ability to design and modularize medium-to-large-scale embedded software architectures.

精通 C 语言嵌入式开发，具备中大型嵌入式软件架构设计与模块化开发能力。

• Familiar with mainstream MCU/DSP development platforms, proficient in mainstream communication protocols such as CAN,

MODBUS, RS485, and PMBus, and skilled in multi-node device networking, data interaction, and protocol adaptation development.

熟悉主流MCU/DSP开发平台， 熟练掌握CAN、MODBUS、RS485、PMBus 等主流通信协议，精通多节点设备组网、数据交互、协议适配开发。

• A deep understanding of PSU control logic and AI power system topology is preferred, along with mastery of PSU control strategies and core algorithms for fault interlocking protection. Familiarity with 380V AC and 800V DC power supply architectures for computing cabinets is also a plus.

深入理解PSU控制逻辑、AI电源系统拓扑结构，掌握PSU控制策略、故障连锁保护核心算法，熟悉 380V AC、800V DC 算力柜供电架构优先。

• Possesses independent troubleshooting and performance optimization capabilities, and is familiar with software stability, real-time performance, and security optimization solutions.

具备独立的问题排查、性能优化能力，熟悉软件稳定性、实时性、安全性优化方案。

• Candidates with experience in developing AI data centers, server rack energy storage, and UPS systems for data centers will be given priority; those familiar with power supply specifications for intelligent computing centers and data center operation and maintenance systems will also be given priority.

有AI 数据中心、服务器机柜储能、机房 UPS 系统开发经验者优先，熟悉智算中心供电规范、机房运维体系优先。

• Good coding style, strong teamwork and documentation skills, and able to adapt to project iteration pace.

代码规范、具备良好的团队协作能力和文档输出能力，能适配项目迭代节奏。