岗位职责：对专门为大规模LLM训练和推理以及NVIDIA超级计算机和分布式系统上的高性能计算（HPC）量身定制的AI工作负载的性能进行基准测试、分析和分析。
为内部销售、营销、软件和硬件团队汇总并生成包含测试数据的书面和视觉报告 使用适当的硬件和软件设置和配置系统以运行基准测试 与内部团队合作，调试和改进性能问题 开发Python脚本以自动化各种应用程序的测试 协助开发工具和流程，以提高我们执行自动化测试的能力 任职要求： 目前正在攻读计算机科学、电气工程或相关领域的学士学位（或更高学位）。
有使用Python或Unix shell等脚本语言进行编程和调试的经验。
强大的数据分析技能和在书面报告中总结调查结果的能力 有基于Linux系统的实践经验。
熟悉使用Docker或Singularity等容器平台。
有从源代码编译和运行软件的经验。

Have you ever wanted to join a groundbreaking company that is crafting the future of the tech industry? Are you a creative individual with an analytical attitude and a real passion for technology? If this sounds like you then we want to hear from you. We are looking for a dedicated team member to be part of our Performance Lab team to build and improve the work class lab. What you’ll be doing:
1、 Candidate will create, tune, run and analyze graphics and systems benchmarks on PCs, workstations and laptops. Tasks may include some scripting to automate current test processes and benchmarks for improved efficiency.
2、 Candidate may also need to configure computer systems with appropriate hardware and software to run benchmarks on various systems. They would run benchmarks on these systems.
3、 Candidate would analyze results from benchmark runs and create reports to position NVIDIA products appropriately using their evaluation. What we need to see:
1、 BS in Computer Science or similar computer discipline.
2、 Working knowledge of all Windows operating systems.
3、 Linux and Windows knowledge.
4、 Good knowledge of PC systems and components.
5、 Good organizational, time management and task prioritization skills. Ways to stand out from the crowd:
1、 Knowledge of C, C++, Perl, OpenGL, Direct X and D3D.
2、 Familiarity with Linux is a plus.
3、 Good knowledge of NVIDIA GeForce and RTX series.
4、 Familiar with Android application is a plus.

Target: 2027年应届毕业生 NVIDIA pioneered accelerated computing to tackle challenges no one else can solve. Our work in AI and digital twins is transforming the world's largest industries and profoundly impacting society — from gaming to robotics, self-driving cars to life-saving healthcare, climate change to virtual worlds where we can all connect and create. Our internships offer an excellent opportunity to expand your career and get hands on experience with one of our industry leading Autonomous Vehicles and Robotics teams. We’re seeking strategic, ambitious, hard-working, and creative individuals who are passionate about helping us tackle challenges no one else can solve. Potential Internships in this field include: Autonomous Vehicles:
1、 Developing and training state-of-the-art Deep Neural Networks for path generation.
2、 Collecting training datasets and real-time inference run-times using simulators/gyms as well as performing in-vehicle tests.
3、 Course or internship experience related to the following areas could be required: Computer Vision, Mapping, Localization, SLAM, Image Processing, Segmentation. Robotics
1、 Building the fundamental infrastructure and software platforms of our system, working at the very heart of the software system, which will power every robot and application built with Isaac.
2、 Course or internship experience related to the following areas could be required: Robotics, Autonomous Vehicles, Validation Frameworks for Deep Learning, Operating Systems and Data Structures (threads, processes, memory, synchronization), Physics Simulation, Simulators, Computer Graphics, Version Control, Computer Vision, Cloud Technologies. What we need to see:
1、 Must be actively enrolled in a university pursuing a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Engineering, or a related field, for the entire duration of the internship.
2、 Depending on the internship role, prior experience or knowledge requirements could include the following programming skills and technologies: C, C++, CUDA, ROS, Python, OpenGL, Linux, Sensor Input Devices (LiDAR, cameras, radars), Training Frameworks (TensorFlow, Keras, PyTorch).

Target: 2027年应届毕业生 At NVIDIA, we're pioneering transformative AI technologies that reshape industries globally. Join our dynamic, forward-thinking team working directly at the bleeding edge of Large Language Models (LLMs). This internship offers a unique opportunity to contribute to ​customer-accelerated deployments​ where your engineering skills directly impact leading enterprises. You'll collaborate closely with senior engineers and hardware experts to push the boundaries of efficiency and capability in LLM training and inference. Gain unparalleled experience applying your knowledge to real-world challenges on NVIDIA's industry-defining Tensor Core GPUs and full-stack AI platforms. Stand shoulder-to-shoulder with world-class researchers and engineers solving the next generation of AI scale and speed. What you’ll be doing
1、 ​Drive LLM efficiency:​​ Design and leverage advanced low-precision quantization techniques (INT8, FP8, FP4) to optimize inference performance for customer deployments.
2、 ​Innovate with frameworks:​​ Simulate, optimize, and extend cutting-edge training & inference frameworks (e.g., vLLM, SGLang, TensorRT-LLM, NeMo, Megatron) within NVIDIA's ecosystem.
3、 ​Enable new AI capabilities:​​ Integrate and validate next-generation LLM architectures and features within core frameworks to expand NVIDIA's solution offerings.
4、 ​Tune for peak performance:​​ Conduct rigorous performance analysis and tuning of LLM workloads for optimal execution on cloud and on-premises NVIDIA platforms.
5、 ​Collaborate on customer solutions:​​ Partner with engineering teams and solution architects to translate customer requirements into high-impact LLM engineering implementations. What we need to see
1、 Pursuing a ​MS or PhD​ in Computer Science, Artificial Intelligence, Electrical Engineering, or a related field.
2、 ​Hands-on experience​ with large language model (LLM) training and/or inference frameworks from project work, research, or prior internships.
3、 ​Strong proficiency in PyTorch​ and ​Python programming.
4、 ​Solid foundational understanding​ of:
1、 Transformer architectures & core LLM algorithms.
2、 Principles and trade-offs of model quantization techniques.
3、 Distributed training paradigms (e.g., FSDP, ZeRO, 3D/5D parallelism, RLHF infrastructure).
4、 A ​link to your GitHub profile or code samples​ is required with your application (demonstrating relevant projects). Ways to stand out from the crowd
1、 Demonstrable experience with ​quantization tools and workflows​ (e.g., GPTQ, AWQ, SmoothQuant).
2、 ​Contributions to relevant Open Source Software projects​ (e.g., vLLM, SGLang, Hugging Face Transformers, PyTorch, DeepSpeed).
3、 ​Understanding of GPU architecture (CUDA), high-performance computing concepts, and cluster communication libraries​ (e.g., NCCL, MPI).
4、 Record of ​published research​ in machine learning, NLP, or systems at major conferences/journals.
5、 Experience deploying or optimizing workloads on ​NVIDIA GPUs​ and familiarity with ​NVIDIA AI software stacks.

Target: 2027年应届毕业生 NVIDIA pioneered accelerated computing to tackle challenges no one else can solve. Our work in AI and digital twins is transforming the world's largest industries and profoundly impacting society — from gaming to robotics, self-driving cars to life-saving healthcare, climate change to virtual worlds where we can all connect and create. Our internships offer an excellent opportunity to expand your career and get hands on experience with one of our industry leading Deep Learning Computer Architecture teams. We’re seeking strategic, ambitious, hard-working, and creative individuals who are passionate about helping us tackle challenges no one else can solve. We’re looking for students pursuing Bachelor's, Master's, or PhD degree within a relevant or related field. What we need to see:
1、 Must be actively enrolled in a university pursuing a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Engineering, or a related field, for the entire duration of the internship.
2、 Course or internship experience related to the following areas could be required:
1、 Computer Architecture experience in one or more of these focus areas: GPU Architecture, CPU Architecture, Deep Learning, GPU Computing, Parallel Programming, or High-Performance Computing Systems.
2、 GPU Computing (CUDA, OpenCL, OpenACC), GPU Memory Systems, Deep Learning Frameworks (PyTorch, TensorFlow, Keras, Caffe), HPC (MPI, OpenMP).
3、Modelling/Performance Analysis, Parallel Processing, Neural Network Architectures, GPU Acceleration, Deep Learning Neural Networks, Compiler Programming.
4、 Performance Modeling, Profiling, Optimizing, and/or Analysis.
5、 Depending on the internship role, prior experience or knowledge requirements could include the following programming skills and technologies: C, C++, Python, Perl, GPU Computing (CUDA, OpenCL, OpenACC), Deep Learning Frameworks (PyTorch, TensorFlow, Caffe), HPC (MPI, OpenMP).

Target: 2027年应届毕业生 NVIDIA pioneered accelerated computing to tackle challenges no one else can solve. Our work in AI and digital twins is transforming the world's largest industries and profoundly impacting society — from gaming to robotics, self-driving cars to life-saving healthcare, climate change to virtual worlds where we can all connect and create. Our internships offer an excellent opportunity to expand your career and get hands on experience with one of our industry leading Computer Architecture teams. We’re seeking strategic, ambitious, hard-working, and creative individuals who are passionate about helping us tackle challenges no one else can solve. We’re looking for students pursuing Bachelor's, Master's, or PhD degree within a relevant or related field. What we need to see:
1、 Must be actively enrolled in a university pursuing a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Engineering, or a related field, for the entire duration of the internship.
2、 Course or internship experience related to the following areas could be required:
1、 Computer Architecture experience in one or more of these focus areas: Computer Graphics, Deep Learning, Ray Tracing, Parallel Programming, Memory Architecture, or High-Performance Computing Systems.
2、 Deep Learning, Modelling/Performance Analysis, Parallel Programming.
3、 Digital Systems, VLSI Design, GPU or CPU Architecture, Computer Arithmetic, CMOS Transistors and Circuits, Computer Organization, Computer System Design.
4、 Hardware Control Programming, GPU Acceleration, Parallel Processing (CUDA), High-Performance Computing (HPC), Embedded Systems.
5、 Depending on the internship role, prior experience or knowledge requirements could include the following programming skills and technologies: Verilog, SystemVerilog, VHDL, Linux, C, C++, Python, Perl, Modern Graphics APIs (DirectX, OpenGL, Vulkan), GPU Computing (CUDA, OpenCL), Revision Control (Perforce, Git), HPC (MPI, OpenMP).

Target: 2027年应届毕业生 NVIDIA pioneered accelerated computing to tackle challenges no one else can solve. Our work in AI and digital twins is transforming the world's largest industries and profoundly impacting society — from gaming to robotics, self-driving cars to life-saving healthcare, climate change to virtual worlds where we can all connect and create. Our internships offer an excellent opportunity to expand your career and get hands on experience with one of our industry leading Hardware teams. We’re seeking strategic, ambitious, hard-working, and creative individuals who are passionate about helping us tackle challenges no one else can solve. What we need to see:
1、 Must be actively enrolled in a university pursuing a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Engineering, or a related field, for the entire duration of the internship.
2、 Course or internship experience related to the following areas could be required:
1、 Digital Systems, Digital Design, VLSI Design, RTL Design, Chip Blueprint, Power and Performance Optimization, Testbench Creation.
2、 Computer Architecture, Computer Arithmetic, Object-Oriented Programming, CMOS Transistors, Circuits, Infrastructure Development.
3、 Architectural Tradeoff Analysis, Power and Noise Analysis, Silicon Instrumentation and Measurement.
4、 CAD and Physical Design Methodologies (Flow and Tools Development), Chop Floorplan, Power/Clock Distribution, Packaging, P&R and Timing Closure.
5、 Performance, Verification, and Emulation Methodology.
6、 Depending on the internship role, prior experience or knowledge requirements could include the following programming skills and technologies: SystemVerilog, Verilog, VHDL, UVM, Perl, Python, TCL, C, C++, Linux, Testbench Tools.

Target: 2027年应届毕业生 NVIDIA pioneered accelerated computing to tackle challenges no one else can solve. Our work in AI and digital twins is transforming the world's largest industries and profoundly impacting society — from gaming to robotics, self-driving cars to life-saving healthcare, climate change to virtual worlds where we can all connect and create. Our internships offer an excellent opportunity to expand your career and get hands on experience with one of our industry leading Software teams. We’re seeking strategic, ambitious, hard-working, and creative individuals who are passionate about helping us tackle challenges no one else can solve. Potential Internships in this field include: Development Tools：
1、 Debugging complex system-level issues using Jenkins.
2、 Course or internship experience related to the following areas could be required: Relational Databases, Linear Algebra & Numerical Methods, Operating Systems (memory/resource management), Scheduling and Process Control, Hardware Virtualization. Cloud:
1、 Supporting overall architecture and design of our cloud storage infrastructure.
2、 Implementing and troubleshooting storage and data platform tools, automating storage infrastructure end-to-end.
3、 Course or internship experience related to the following areas could be required: Distributed Systems, Data Structures & Algorithms, Virtualization, Automation/ scripting, Container & Cluster Management, Debugging. Tools Infrastructure:
1、 Building industry leading technology by proving workflows and infrastructure, alongside a team of experts in production software development and chip design methodologies.
2、 Enabling success for content running on the chip from application tracing and analysis to modeling, diagnostics, performance tuning, and debugging.
3、 Course or internship experience related to the following areas and technologies could be required: Unix/Shell scripting, Linux, Java, Java script (including Node, React, Vue), C++, CUDA, OOP, Go, Python, Git, GitLab, Perforce, Kubernetes and Microservices, Schedulers (LSF, SLURM), Containers (Docker), Configuration Automation (Ansible). Machine Learning Operations:
1、 Deep Learning, GPU Computing, Accelerated Computing.
2、 Validation Frameworks for Deep Learning, Deep Learning Frameworks and Libraries (NumPy, SciPy, cuBLAS, cuDNN).
3、 Data Preprocessing, Training Acceleration (CUDA, cuDNN, NCCL), Convolution Operations (cuDNN), Real-Time Inference (TensorRT).
4、 Building Infrastructure for Back-End Analytics. What we need to see:
1、 Must be actively enrolled in a university pursuing a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Engineering, or a related field, for the entire duration of the internship.
2、 Depending on the internship role, prior experience or knowledge requirements could include the following programming skills and technologies: Java, Java script, (including Node, React, Vue), SQL, C, C++, CUDA, OOP, Go, Python, Git, Perforce, Kubernetes and Microservices, Schedulers (LSF, SLURM), Containers (Docker), Configuration Automation (Ansible).

Target: 2027年应届毕业生 NVIDIA pioneered accelerated computing to tackle challenges no one else can solve. Our work in AI and digital twins is transforming the world's largest industries and profoundly impacting society — from gaming to robotics, self-driving cars to life-saving healthcare, climate change to virtual worlds where we can all connect and create. Our internships offer an excellent opportunity to expand your career and get hands on experience with one of our industry leading Hardware teams. We’re seeking strategic, ambitious, hard-working, and creative individuals who are passionate about helping us tackle challenges no one else can solve. What we need to see:
1、 Must be actively enrolled in a university pursuing a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Engineering, or a related field, for the entire duration of the internship.
2、 Course or internship experience related to the following areas could be required:
1、 Formal Verification, GPU or processor Verification or Validation.
2、 Digital Systems, Digital Design, VLSI Design, RTL Design, Testbench Creation.
3、 Random functional testing, writing test plans, directed/random diagnostics.
4、 CPU and GPU Architecture, Computer Architecture, Software Infrastructure (for validation of hardware architecture).
5、 Depending on the internship role, prior experience or knowledge requirements could include the following programming skills and technologies: Verilog, SystemVerilog, VHDL, UVM, Perl, TCL, C, C++, Linux, Testbench Tools.

岗位职责： LLM inference • 大模型推理系统优化:通过系统手段来提升TensorRT-LLM的性能、易用性和功能丰富度。
包括但不限于TensorRT-LLM的API改进、LLM runtime优化、服务层优化、定制优化算子、分布式优化等等。
• 大模型推理模型优化: 通过模型优化的手段进一步提升TensorRT-LLM的性能表现，包括低比特量化，KV Cache量化，稀疏化等等，也包括类似于Speculative decoding, Streaming-LLM这类模型系统联合优化的工作内容。
AI Complier System • AI编译器: 主要包括如何使用AI编译器技术，在NV的当代硬件，以及未来硬件上，以自动化的方式生成高效的计算密集/访存密集的算子kernel。
岗位要求：
1、有大模型推理优化相关经验
2、可以实习3个月左右，一周至少出勤3天
3、提供转正机会

我们现诚聘 VLSI 物理设计实习生！27届同学提供转正机会 NVIDIA 中国 VLSI 物理设计团队成立于 2005 年，30 多年来为 NVIDIA 公司推出的众多成功产品做出了贡献。
我们运用最新的工艺技术、先进的 EDA 工具和精湛的设计方法，始终致力于最具挑战性的设计，并不断突破性能极限。
工作内容：
1、负责内部流程的优化，包括布局规划、电源/时钟分配、布局布线、时序/功耗/噪声分析、芯片组装和物理验证。
2、负责内部工具的开发和探索。
3、寻找 AI 解决方案，提升工作流程效率和自动化程度。
职位要求：
1、电子工程、计算机科学和数学/物理学硕士学位。
2、具备 IC 设计或算法及编程知识。
脱颖而出的秘诀：
1、IC 设计或物理设计实践经验。
2、常用编程语言（例如 Python）的实践经验。
3、深度学习/机器学习项目经验。

我们现诚聘 VLSI 物理设计实习生！27届同学提供转正机会 NVIDIA 中国 VLSI 物理设计团队成立于 2005 年，30 多年来为 NVIDIA 公司推出的众多成功产品做出了贡献。
我们运用最新的工艺技术、先进的 EDA 工具和精湛的设计方法，始终致力于最具挑战性的设计，并不断突破性能极限。
工作内容：
1、负责内部流程的优化，包括布局规划、电源/时钟分配、布局布线、时序/功耗/噪声分析、芯片组装和物理验证。
2、负责内部工具的开发和探索。
3、寻找 AI 解决方案，提升工作流程效率和自动化程度。
职位要求：
1、电子工程、计算机科学和数学/物理学硕士学位。
2、具备 IC 设计或算法及编程知识。
脱颖而出的秘诀：
1、IC 设计或物理设计实践经验。
2、常用编程语言（例如 Python）的实践经验。
3、深度学习/机器学习项目经验。

岗位职责： LLM inference • 大模型推理系统优化:通过系统手段来提升TensorRT-LLM的性能、易用性和功能丰富度。
包括但不限于TensorRT-LLM的API改进、LLM runtime优化、服务层优化、定制优化算子、分布式优化等等。
• 大模型推理模型优化: 通过模型优化的手段进一步提升TensorRT-LLM的性能表现，包括低比特量化，KV Cache量化，稀疏化等等，也包括类似于Speculative decoding, Streaming-LLM这类模型系统联合优化的工作内容。
AI Complier System • AI编译器: 主要包括如何使用AI编译器技术，在NV的当代硬件，以及未来硬件上，以自动化的方式生成高效的计算密集/访存密集的算子kernel。
岗位要求：
1、有大模型推理优化相关经验
2、可以实习3个月左右，一周至少出勤3天
3、提供转正机会

岗位职责： 研究和实现用于游戏引擎和创作工具的创新生成式人工智能，包括实时神经图形、基于物理的动画和扩散模型。 开发神经图形、动画和物理模型，并维护用于游戏制作和用户运行时的开源项目。将它们集成到主流游戏引擎和DCC工具中。 任职要求： 27届在校生，可实习4-6个月 有机器人、游戏引擎方向的经验，对游戏感兴趣

岗位职责： 研究和实现用于游戏引擎和创作工具的创新生成式人工智能，包括实时神经图形、基于物理的动画和扩散模型。 开发神经图形、动画和物理模型，并维护用于游戏制作和用户运行时的开源项目。将它们集成到主流游戏引擎和DCC工具中。 任职要求： 27届在校生，可实习4-6个月 有机器人、游戏引擎方向的经验，对游戏感兴趣 要求有对刚体和关节的物理模拟有深入的理解。

工作内容： 开发和优化控制系统，包括运动、操作和全身控制算法； 在物理仿真和真实类人硬件上部署和评估神经网络模型； 设计并维护低延迟、高精度的类人机器人远程操作软件； 实施工具和流程以定期进行机器人维护、诊断和故障排除，确保系统可靠性； 监控实验室的远程操作员，开发质量保证工作流程，确保高质量数据收集； 与研究人员合作进行模型训练、数据处理和机器学习运营周期。 任职要求： 在计算机科学、机器人、工程或相关领域攻读学士学位； 有在机器人硬件上部署和调试神经网络模型的实践经验； 能够实现实时控制算法、远程操作系统和传感器融合； 精通 Python、C++ 等语言，对机器人框架（ROS）和物理仿真（Gazebo、Mujoco、Isaac 等）有经验； 有维护和故障排除机器人系统（包括机械、电气和软件组件）的经验； 必须在上海现场工作。

该岗位为春季实习生岗位，面试环节将于11月后陆续开始，期望候选人能于2026年2月后入职（具体时间可以具体沟通），欢迎实习时间合适的同学踊跃投递！ 软件测试开发组介绍： 软件测试开发团队负责测试计划、执行和报告，编写脚本以自动化测试，为QA团队设计和开发工具，开发用于验证的集成测试，识别弱点，不断设计更好、更有创意的测试计划。软件测试开发团队对NVIDIA产品的质量有极大的意义和影响。 该岗位有不同方向，包含：微服务、深度学习、CUDA测试开发、大数据、GPU测试开发 等等，具体方向HR将根据简历进行推荐和流转。 任职要求： - 27届同学欢迎投递 - 预计春节后入职，可在上海长期线下实习（12个月） - 计算机/软件工程等相关专业本科及以上学历 - 有软件测试开发的相关实习、项目经历 - 对软件测试开发工作感兴趣、有热情

该岗位为春季实习生岗位，面试环节将于11月后陆续开始，期望候选人能于2026年2月后入职（具体时间可以具体沟通），欢迎实习时间合适的同学踊跃投递！ By submitting your resume, you’re expressing interest in one of our 2026 Hardware ASIC Design Internships. We’ll review resumes on an ongoing basis, and a recruiter may reach out if your experience fits one of our many internship opportunities. NVIDIA pioneered accelerated computing to tackle challenges no one else can solve. Our work in AI and digital twins is transforming the world's largest industries and profoundly impacting society — from gaming to robotics, self-driving cars to life-saving healthcare, climate change to virtual worlds where we can all connect and create. Our internships offer an excellent opportunity to expand your career and get hands on experience with one of our industry leading Hardware teams. We’re seeking strategic, ambitious, hard-working, and creative individuals who are passionate about helping us tackle challenges no one else can solve. What we need to see: Must be actively enrolled in a university pursuing a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Engineering, or a related field, for the entire duration of the internship. Course or internship experience related to the following areas could be required: Digital Systems, Digital Design, VLSI Design, RTL Design, Chip Blueprint, Power and Performance Optimization, Testbench Creation Computer Architecture, Computer Arithmetic, Object-Oriented Programming, CMOS Transistors, Circuits, Infrastructure Development Architectural Tradeoff Analysis, Power and Noise Analysis, Silicon Instrumentation and Measurement CAD and Physical Design Methodologies (Flow and Tools Development), Chop Floorplan, Power/Clock Distribution, Packaging, P&R and Timing Closure Performance, Verification, and Emulation Methodology Depending on the internship role, prior experience or knowledge requirements could include the following programming skills and technologies: SystemVerilog, Verilog, VHDL, UVM, Perl, Python, TCL, C, C++, Linux, Testbench Tools

该岗位为春季实习生岗位，面试环节将于11月后陆续开始，期望候选人能于2026年2月后入职（具体时间可以具体沟通），欢迎实习时间合适的同学踊跃投递！ VLSI Physical Design Team at NVIDIA China has been built up since 2005. The team has made contribution to various successful products launched by NVIDIA Corporation over 30 years. We utilize latest process technology, advanced EDA tools, and sophisticated design methodology. We always work on the most challenging designs, and push for performance limit. What you will be doing: Work on flow enhancement of in-house flow on floorplan, power/clock distribution, placement, routing, timing/power/noise analysis, chip assembly, and physical verification Work on in-house tool development and exploration Look for AI solution to improve workflow efficiency and automation What we need to see: MS in EE, CS and Mathematics/Physics Knowledge of IC design or algorithm and programming Ways to stand out from the crowd: Hands-on experience in IC design or physical design Hands-on experience in popular programming language (Python etc.) DL/ML project experience

注：Target 27年毕业生，要求春节后入职，线下实习3-6个月 What you'll be doing: 1. Develop and refine software solutions to expedite LLM SW stack (could be within inference/post train or pre-train phase) by harnessing the power of GPU technology. 2. Collaborate closely with a world-class team of engineers to implement and refine GPU-based algorithms. 3. Analyze and determine the most effective methods to improve performance, ensuring seamless execution across diverse computing environments. 4. Engage in both individual and team projects, contributing to NVIDIA's mission of leading the AI revolution. 5. Work in an empowering and inclusive environment to successfully implement groundbreaking AI solutions. What we need to see: 1. Proven experience in software engineering, particularly in GPU programming and LLM inference. 2. Strong proficiency in programming languages such as Python, C++, and CUDA. 3. A solid understanding of deep learning frameworks and techniques. 4. Outstanding problem-solving skills and the ability to work collaboratively in a team setting. 5. Ambitious approach with a proven track record of taking initiative and delivering results. 6. A degree in Computer Science, Engineering, or a related field, or equivalent experience. 7. Experience with PyTorch, RLHF (Reinforcement Learning with Human Feedback), and LLM training frameworks like FSDP/Megatron LLM is a plus.

该岗位为春季实习生岗位，面试环节将于11月后陆续开始，期望候选人能于2026年2月后入职（具体时间可以具体沟通），欢迎实习时间合适的同学踊跃投递！ VLSI Physical Design Team at NVIDIA China has been built up since 2005. The team has made contribution to various successful products launched by NVIDIA Corporation over 30 years. We utilize latest process technology, advanced EDA tools, and sophisticated design methodology. We always work on the most challenging designs, and push for performance limit. What you will be doing: Work on flow enhancement of in-house flow on floorplan, power/clock distribution, placement, routing, timing/power/noise analysis, chip assembly, and physical verification Work on in-house tool development and exploration Look for AI solution to improve workflow efficiency and automation What we need to see: MS in EE, CS and Mathematics/Physics Knowledge of IC design or algorithm and programming Ways to stand out from the crowd: Hands-on experience in IC design or physical design Hands-on experience in popular programming language (Python etc.) DL/ML project experience

What you’ll be doing: 1. Investigate and propose architecture ideas based on quantitative study of existing and projected SM architecture. 2. Develop performance and functional simulation models. 3. Develop performance and functional test plan and tests to validate new SM architectural and features. 4. Test and debug on simulators, RTL and real silicon. What we need to see: 1. Good mastery of C++ language. 2. Experience of performance/functional modelling, profiling and analysis is a plus. 3. Practice with trace-driven and execution-driven simulation model development is a plus. 4. Pursuing BSEE or BSCSE. MS or PhD is a plus. Ways to stand out from the crowd: 1. Candidates with GPU shader unit design, microprocessor design, and CPU/GPU performance analysis experience are preferred.

What You'll Be Doing: 1. Physical layout of STD cells, ROMs, and compiled RAMs on each most advanced foundry processes with earlier and unstable design environments. 2. Work closely with NVIDIA talents to speed up layout production including customized DRC/LVS, and plenty of specific verification development, testing, and debug. What We Need To See: 1. Pursuing MS or PH.D. in MS Microelectronics, Semiconductor Physics or related field. 2. You are familiar with Cadence Virtuoso, you're familiar with Siemens Calibre or Synopsis ICV, or similar design and verification tools. 3. Experience working across teams and dealing the tasks in parallel. 4. Excellent communication, problem-solving, and analytical skills to decompose complex issues and present them clearly and simply. Ways To Stand Out From The Crowd: 1. Knowledge of and experience with advanced FinFET/GAA processes. 2. Prior design experience involving Standard Cells and/or Memory structures. 3. Good English written, verbal, and presentation skills. 4. scripting ability in SKILL and/or Perl.

注：该岗位Target 27年毕业生 What you will be doing: 1. Working with architects, design leads, physical design leads and package leads, you will develop and to craft and optimize floorplans during early chip development. 2. Drive the area review process and collaborate with the ASIC design team to identify area, interconnect and floorplan improvement opportunities. 3. Solve timing and routing congestion issues with physical and ASIC design teams by influencing early design and physical implementation decisions. 4. You will build tools and improve existing infrastructure using AI extensively to optimize chip area and speed of execution. What we need to see: 1. Master‘s Degree in Electrical Engineering, Computer Science, or Computer Engineering or equivalent work experience. 2. A deep hardware engineering background with a concentration in VLSI and/or Computer Architecture. 3. Experience in Verilog, System Verilog or similar HVL. 4. Experience with CAD and physical design methodologies (flow and tool development), chip floorplan, power/clock distribution, packaging, P&R and timing closure. 5. Strong communication and interpersonal skills and ability & desire to work as a great teammate should be displayed in your interview. 6. Python, Perl and C/C++ programming language experience. 7. Experience using AI coding assistants like Cursor or Github copilot or Windchill. Ways to stand out from the crowd: 1. Experience in driving development of large-scale ASIC floorplan is a huge plus. 2. Experience building large projects from the ground up using AI tools. Good command of spoken English.

注：该岗位Target 27年毕业生 该岗位工作地点可选择上海或北京 What you will be doing: 1. Work on flow enhancement of in-house flow on floorplan, power/clock distribution, placement, routing, timing/power/noise analysis, chip assembly, and physical verification. 2. Work on in-house tool development and exploration. 3. Look for AI solution to improve workflow efficiency and automation. What we need to see: 1. Pursuing MS in EE, CS and Mathematics/Physics. 2. Knowledge of IC design or algorithm and programming. Ways to stand out from the crowd: 1. Hands-on experience in IC design or physical design. 2. Hands-on experience in popular programming language (Python etc.). 3. DL/ML project experience.

入职时间：2026年2月及以后 实习时间要求：3-6个月及以上 NVIDIA开发工具团队正在寻找一名软件工程师实习生加入我们的工作，以推进图形和计算性能分析和调优的状态。您将帮助汽车、VR、游戏、深度学习和高性能计算领域突破性产品的开发人员分析和改进其产品的性能。您将有机会学习世界上最复杂的GPU的流水线和驱动程序堆栈，与来自世界各地的一群才华横溢的工程师合作，并运用您的软件开发技能来改进我们的产品。 任职要求： -可以线下实习3-6个月或以上 -精通C/C++ -熟悉Graphics API/CUDA -英文沟通顺畅

岗位职责: 1. 负责芯片集成和网表（netlist）生成，并跨团队协作以实现芯片分区（partitioning）和布局规划（floorplan）。 2. 进行逻辑综合（Synthesis）、RTL/网表质量检查、形式验证（formal verification）以及功能 ECO（Engineering Change Order）的创建。 3. 负责时序约束（constraints）的创建与验证、时序预算（timing budget）的制定，与 ASIC 团队合作分析并解决功能时序问题，完成所有特殊路径的时序收敛（timing closure），如 IO、测试、时钟、异步路径等。 4. 与布局布线（Place & Route）工程师协作进行芯片物理实现，以达成全芯片时序收敛。 5. 开发和改进从前端（布局前）到后端（布局后）的完整时序收敛流程，并为上述领域开发自动化流程。 6. 参与上述任一领域的方法学（methodology）研究与开发。 任职要求: 1. 相关专业领域的在读硕士或博士。 2. 具备扎实的集成电路（IC）设计知识。 3. 对技术研究和 IC 后端设计充满热情。 4. 具备自我驱动力、良好的学习能力和团队合作精神。 5. 优秀的沟通技巧，以及熟练的英语读写能力。 优先条件: 1. 有 IC 后端实现经验者优先。 2. 具备 Synopsys（DC/PT/Formality）、Cadence（RC compiler/LEC）等 EDA 软件的实践经验者优先。 3. 熟练使用 Python 或 Tcl 脚本语言者优先。

岗位职责： 1. 负责 DFT 逻辑的实现与验证。 2. 维护 DFT 流程并运行回归测试。 3. 运行 DFT 验证。 4. 负责测试向量（pattern）的生成与调试。 任职要求： 1. 具备扎实的逻辑设计与验证背景。 2. 掌握基本的 DFT 知识。 3. 熟练掌握 Perl/Tcl 编程。 4. 精通 Verilog HDL。 5. 电子工程（EE）相关专业的在读硕士或博士。 优先条件 1. 有逻辑仿真器和调试工具（如 vcs、ncsim、verdi）使用经验者优先。 2. 熟悉 C/C++、Makefile 者为佳。 3. 具备出色的问题解决和分析能力。 4. 优秀的沟通能力和团队合作精神。

岗位职责： 1. 使用公司内部开发的工具和行业标准的硅前（pre-silicon）门级与 RTL 功耗分析工具，帮助提升产品能效。 2. 开发并分享执行硅前功耗分析的最佳实践，优化内部功耗工具并实现最佳实践的自动化。 3. 执行对比功耗分析，以发现需要进一步研究的趋势和异常。 4. 与架构师和 RTL 设计师沟通协作，帮助他们解读功耗数据、识别功耗缺陷（power bugs），并推动实施修复方案。 5. 为功耗分析选择并运行多种工作负载，与性能和架构团队合作，验证工作负载的性能。 6. 使用 Verilog 语言为新的架构特性构建原型并进行功耗分析。 任职要求： 1. 相关专业领域的在读硕士或博士。 2. 对能耗、功耗估算和低功耗设计等概念有基本了解。 3. 熟悉 Verilog 语言和 ASIC 设计原则，了解逻辑单元（logic cells）相关知识。 4. 具备良好的英语口头和书面沟通能力及人际交往技巧，该职位需要与设计团队进行大量协作。 5. 具备扎实的编程能力，优先考虑精通 Python、C++ 的候选人。 6. 能够构建和分析算法，并评估其时间复杂度和内存消耗。 7. 热衷于运用数据驱动决策和分析方法来改进我们的产品。 优先条件： 熟悉功耗分析工具及流程开发者将作为重要的加分项。

岗位职责： 1. 致力于内部流程的优化，涵盖布局规划（floorplan）、电源/时钟分配、布局（placement）、布线（routing）、时序/功耗/噪声分析、芯片整合以及物理验证等环节。 2. 参与内部工具的开发与探索。 3. 探索人工智能解决方案，以提升工作流程的效率和自动化水平。 任职要求： 1. 电子工程（EE）、计算机科学（CS）、数学或物理学相关专业的在读硕士。 2. 具备集成电路（IC）设计或算法与编程方面的知识。 优先条件 1. 具备集成电路设计或物理设计的实践经验者优先。 2. 具备主流编程语言（如 Python）的实践经验者优先。 3. 有深度学习/机器学习（DL/ML）项目经验者优先。

该岗位需要每周能够工作 4 天，希望有1年可实习时间。 我们现招聘一名深度学习软件测试开发工程实习生。 该职位隶属于 NVIDIA 深度学习软件质量保证团队，负责定义、开发和执行测试，以验证 NVIDIA 深度学习软件和 GPU 基础架构的稳健性，并衡量其在自动驾驶、医疗保健、语音识别、自然语言处理以及其他各种 AI 场景下的性能。该团队与多个 AI 产品团队合作，开发新产品；制定并改进复杂的测试计划；并改进我们针对各种 GPU 计算平台的工作流程。您需要在关键路径上成长，为价值数十亿美元的业务线的开发人员提供支持，并深刻理解响应能力、严谨性和团队合作的价值。您需要在您的领域不断促进和实施效率改进。加入我们，共同构建全球通用的软件！ 工作职责： • 负责 NVIDIA 深度学习软件产品（例如 cuDNN、TensorRT、NVIDIA 优化框架（例如 TensorFlow、PyTorch、MxNET 等））的 GPU 软件测试和测试自动化改进。 • 负责深度学习软件栈发布中的功能性、兼容性和性能测试。 • 与开发团队合作，对问题进行分类、根本原因分析、验证修复方案、定义新测试并改进测试计划。 • 利用 AI 驱动的工具提高效率和质量，包括测试用例/计划/脚本生成、缺陷检测、CBTP、错误修复和日常支持。 我们需要看到： • 计算机科学/电子工程/计算机工程学士或更高学位。 • 精通脚本语言（Python、Perl、bash），并具备 Linux 知识。 • 具有 C/C++ 编程经验者优先。 • 熟悉任何深度学习框架者优先。 • 良好的沟通能力，流利的英语口语和书面表达能力。 • 拥有 AI 工具使用经验。 脱颖而出的秘诀： • 熟悉 NVIDIA GPU 硬件是一大优势。 • 拥有 NVIDIA GPU 计算 (CUDA) 背景是一大优势。已成功利用 AI 工具显著提高效率、简化工作流程或增强流程自动化。

What you will be doing: 1. Work proactively on GPU/SOC feature/IP bring-up and characterization including creating validation, tuning and optimization methodologies, develop proper test plan and test cases. 2. Perform system level use case analysis/profiling and feature return-on-investment investigations, prototype and validation, design methods and control policies to bring the new technology into production and transcend product goals. 3. Collaborate across System Architecture, DFT, ASIC, SW/FW, platform, validation, and production teams throughout the product life cycle on system-level architecture, design, productization, debugging, and deployment for complex silicon designs to improve quality, safety, and manufacturability. 4. Design tools/ script to automate product definitions, data collection, test case execution, and results analysis. Provide detailed data analysis of functionality, performance, and latency. 5. Hands on actions on silicon bring-up, validation, and debug; Coordinate product level feature deployment to achieve high product quality at aggressive schedule. What we need to see: 1. BS or MS degree in Microelectronics/Electronic Engineering/Computer and Related Professionals. 2. EE fundamentals, knowledgeable in digital design, computer architecture, power analysis, timing analysis, fault analysis, sampling, statistics, and scripting. 3. Experience in one or more areas of high-performance circuit (Low power, PLLs and clocking, SerDes and high-speed IO, SRAMs, power delivery/regulation, security circuits etc). 4. Good understanding of firmware/driver structures and their interaction with HW is a plus. 5. Experience with Python script development, Windows and Linux OS is a plus. 6. Hands-on lab experience with silicon bring up, frequency and power characterization, Tester to System correlation, lab tools (oscilloscopes, multimeters, DAQ). 7. Excellent problem solving, collaborative, and interpersonal skills. 8. Excellent verbal communication and written, presentation skills in both Chinese and English. 9. Effective in a collaborative environment.

What you'll be doing: 1. Participate in verification of multimedia IP at unit or SoC level. 2. Develop test plans, create testbenches, and write test cases using SystemVerilog/C++/Python. 3. Apply simulation-based or formal verification methodologies to ensure design correctness. 4. Write and maintain verification scripts to automate test execution and regression. 5. Debug failures, analyze coverage, and work closely with designers to resolve issues. 6. Contribute to continuous improvement of verification methodologies and flow. What we need to see: 1. Master’s degree in Electrical Engineering, Computer Engineering, or related field. 2. Strong understanding of digital design fundamentals. 3. Proficiency in System Verilog for verification. 4. Solid programming/ scripting skills (e.g., Python, C++). 5. Knowledge of hardware verification methodologies, especially UVM. 6. Exposure to formal verification tools. 7. Good communication skills in English, both written and spoken. 8. Strong problem-solving skills, self-motivated, and eager to learn.

职位描述： 我们的MMPLEX IP覆盖多个方向，包括视频/显示/安全/加速器/处理器。这些IP几乎存在于所有NVIDIA产品中。MMPLEX验证团队专注于全面的IP验证，从单元级验证到SoC级集成，以及应用先进的形式化验证技术。 你将要做的事情： 1. 参与单元或SoC级多媒体IP的验证。 2. 开发测试计划，创建测试台，并使用SystemVerilog/C++/Python编写测试用例。 3. 应用基于仿真或形式化的验证方法来确保设计的正确性。 4. 编写和维护验证脚本，以自动化测试执行和回归。 5. 调试故障，分析覆盖率，并与设计人员密切合作解决问题。 6. 为验证方法和流程的持续改进做出贡献。 我们需要看到的是： 1. 电气工程、计算机工程或相关专业硕士学位。 2. 对数字设计基础有很强的理解。 3. 熟练掌握SystemVerilog进行验证。 4. 扎实的编程/脚本技能（例如Python、C++）。 5. 了解硬件验证方法，尤其是UVM。 6. 接触正式验证工具。 7. 良好的英语沟通技巧，包括书面和口语。 8. 较强的解决问题的能力，自我激励，渴望学习。

We are now looking for a Power Methodology and Analysis Engineer. NVIDIA prides ourselves in having energy efficient products. We believe that continuing to maintain our products' energy efficiency compared to competition is key to our continued success. Our team is responsible for researching, developing, and deploying methodologies to help NVIDIA's products become more energy efficient; and responsible for analyzing fullchip and unit-level power data, and driving ASIC teams to improve their units’ power efficiency. Key responsibilities include developing techniques to model, analyze, and reduce power consumption of NVIDIA GPUs. As a member of the Power Modeling, Methodology and Analysis Team, you will collaborate with Architects, Performance Engineers, Software Engineers, ASIC Design Engineers, and Physical Design teams to study and implement power analysis and reduction techniques for NVIDIA's next generation GPUs and Tegra SOCs. Your contributions will help us gain early insight into energy consumption of graphics and artificial intelligence workloads, and will allow us to influence architectural, design, and power management improvements. What you'll be doing: Use internally developed tools and industry standard pre-silicon gate-level and RTL power analysis tools, to help improve product power efficiency. Develop and share best practices for performing pre-silicon power analysis, Enhance internal power tools and automate best practices Perform comparative power analysis, to spot trends and anomalies, that warrant more scrutiny. Interact with architects and RTL designers to help them interpret their power data and identify power bugs; drive them to implement fixes. Select and run a wide variety of workloads for power analysis, Collaborate with performance and architecture teams to validate performance of the workloads Prototype a new architectural feature in Verilog and analyze power. What we need to see: EE, MS or PhD in related fields, or equivalent experience. Basic understanding of concepts of energy consumption, estimation, and low power design. Familiarity with Verilog and ASIC design principles, including knowledge of logic cells. Good verbal/written English and interpersonal skills; much collaboration with design teams is expected. Strong coding skills, preferably in Python, C++. Ability to formulate and analyze algorithms, and comment on their time complexity and memory consumption. Desire to bring data-driven decision-making and analytics to improve our products. Ways to s

鼠鼠求职首页