
Huawei envisions a world where technology connects people, empowers industries, and unlocks human potential. Guided by its mission to enrich lives through communication and intelligent innovation, Huawei stands at the forefront of global digital transformation. As a leader in Information and Communications Technology (ICT), the company pioneers breakthroughs in artificial intelligence, cloud computing, and smart devices—building the intelligent foundation of a fully connected world.
Through its Carrier, Enterprise, and Consumer business groups, Huawei delivers resilient digital infrastructure, advanced cloud and AI platforms, and transformative devices that enable progress at every level. Supporting 45 of the world’s top 50 telecom operators and serving one-third of the global population across more than 170 countries, Huawei is shaping a future where connectivity becomes a powerful catalyst for opportunity and sustainable growth.
This spirit of bold innovation is embodied by Huawei Technologies Switzerland AG. From its research hubs in Zurich and Lausanne, pioneering teams push the boundaries of High-Performance Computing, Computer Architecture, Computer Vision, Robotics, Artificial Intelligence, Neuromorphic Computing, Wireless Technologies, and Networking—architecting the intelligent systems that will define tomorrow’s digital era.
The Frontier Systems and Architecture (FroSA) Lab in Zurich researches next-generation computing infrastructure, from the chip to the SuperPOD. Within the lab, the Agentic CPU thrust rethinks the processor core for the era of AI agents. While models execute on accelerators, the CPU hosts everything around them: tool calls, sandboxed code execution, orchestration logic, retrieval, data serialization, and context management. These workloads are bursty, latency-sensitive, and highly concurrent, with large and rapidly changing instruction and data footprints, a poor fit for cores tuned for traditional benchmarks. You will characterize these workloads and invent the microarchitecture that runs them efficiently, from the frontend to the memory hierarchy, helping shape a new research area from its start.
· Agentic Workload Characterization: Build and analyze representative agentic workloads (tool calling, code-execution sandboxes, orchestration runtimes) on real hardware and in simulation to identify the microarchitectural bottlenecks that matter.
· CPU Core Microarchitecture: Design new mechanisms across the core: the frontend (branch prediction, BTB, instruction prefetching, µop caches), register files, instruction scheduling, issue and wake-up logic, out-of-order execution, and scalar/vector/matrix functional units.
· Multithreading and Concurrency: Explore SMT designs, high thread-level concurrency, fast context switching, and mechanisms that reduce resource contention and thrashing among co-running agents.
· Memory Hierarchy: Investigate TLBs, cache hierarchies, prefetching, and the memory subsystem to improve bandwidth, latency, and efficiency for large, irregular footprints.
· Modelling and Prototyping: Evaluate ideas through cycle-level simulation, analytical modelling, profiling on real silicon, and AI-assisted design-space exploration.
· Research Impact: Publish at top-tier venues (ISCA, MICRO, ASPLOS, HPCA), file patents, transfer results to product teams, collaborate with academic partners, and mentor junior researchers.
· PhD in Computer Science, Electrical Engineering, or a related field (MSc considered for the Principal level only).
· Strong background in computer architecture, with deep expertise in CPU microarchitecture.
· Research experience or strong knowledge in one or more of the following areas:
o CPU Microarchitecture: Modern x86 and Arm cores, out-of-order execution, branch prediction, instruction scheduling, and SMT.
o Architecture Simulators: gem5, ChampSim, Sniper, ZSim, QFlex, or similar.
o Vector and Matrix Extensions: Arm SVE/SME or Intel AVX/AMX.
o Memory Systems: Cache hierarchies, TLBs, prefetching, and memory-level parallelism.
o Performance Characterization: Profiling of real systems, top-down bottleneck analysis, and Performance Monitoring Units (PMUs).
· Publication record at top architecture venues (ISCA, MICRO, ASPLOS, HPCA) or equivalent industry impact.
· For Senior Researcher: Owns a research topic end to end, from workload characterization and idea to simulation results and publication. Requires a PhD with several papers at top architecture venues; 2+ years of post-PhD research or industry experience is a plus.
· For Principal Researcher: Defines and leads a research direction spanning several topics, sets the technical agenda with lab leadership, drives the transfer of results into product roadmaps, represents the lab in the research community, and mentors a small team. Requires a PhD plus 5+ years of experience, or an MSc plus 8+ years, with recognized contributions to CPU microarchitecture.
· Strong C/C++ and Python skills for building simulators, models, and analysis tools.
· Excellent analytical and problem-solving skills, clear communication, and the ability to work across research, product, and academic teams.
· Industry experience in CPU performance modelling, design, or validation.
· Knowledge of the software stacks agents run on: language runtimes and JITs (e.g., Python, JavaScript), containers and sandboxes, and OS scheduling.
· Familiarity with agentic AI frameworks and LLM serving systems (e.g., vLLM, SGLang).
· Knowledge of GPU/NPU microarchitecture and heterogeneous CPU–accelerator systems.
· Experience with CUDA, PTX/SASS, or accelerator performance analysis.
· Experience with RTL design (e.g., Verilog, Chisel) or FPGA prototyping.
Don’t tick every box? If you are excited about rethinking the CPU for the agentic era, we would still like to hear from you.
· A New Field to Define: CPU design for agentic workloads is still largely unexplored. The work done now will set its direction, and you will be among the people setting it.
· Research Freedom: Choose ambitious problems, publish at the top architecture venues, and work closely with leading universities.
· Ideas That Reach Silicon: Work with product teams so that the most promising mechanisms can move from simulation toward real processors.
· A Full-Stack Team: Join colleagues spanning CPUs, NPUs, memory systems, and scale-up fabrics, from chip to SuperPOD, a rare setting for true cross-stack co-design.
· Resources to Move Fast: Access substantial compute, simulation infrastructure, and real hardware for profiling.
· Room to Grow: Lead your own topics, mentor junior researchers, and grow toward leading a research direction.
· Life in Zurich: A competitive salary and a high quality of life in one of Europe’s leading hubs for computer architecture research.

Huawei is a leading global provider of information and communications technology (ICT) infrastructure and smart devices. With integrated solutions across four key domains – telecom networks, IT, smart devices, and cloud services – we are committed to bringing digital to every person, home and organization for a fully connected, intelligent world.
Huawei's end-to-end portfolio of products, solutions and services are both competitive and secure. Through open collaboration with ecosystem partners, we create lasting value for our customers, working to empower people, enrich home life, and inspire innovation in organizations of all shapes and sizes.
At Huawei, innovation focuses on customer needs. We invest heavily in basic research, concentrating on technological breakthroughs that drive the world forward. We have more than 207,000 employees, and we operate in more than 170 countries and regions. Founded in 1987, Huawei is a private company fully owned by its employees.
House Rules
This page is for ICT professionals with an interest in Huawei and our industry to engage in open discussions.
To facilitate dialogue, please follow these rules:
- Huawei holds the right to delete comments that are offensive, misleading, false, unlawful, off-topic and in violation of any regulations.
- Repeated violations of any of the above will be removed and users may be blocked.
- Huawei does not necessarily endorse the information shared by members.
- Please be familiar with and follow LinkedIn's User Agreement.
- By publicly uploading a photograph or comment, you give Huawei permission to feature your content. This will always be credited.
Please visit the below portals for career or customer service queries.
Career page: http://bit.ly/2rdljD7
Customer service: http://bit.ly/2a4mXNY
Thank you for visiting us & we hope you enjoy your time on our page.