Research Scientist - AI Performance Optimization and Systems

About Us

Huawei is a global leader in information and communications technology (ICT), driving innovation and shaping the future of an intelligent world. With a presence in over 170 countries and a strong commitment to research and development, Huawei continues to push the boundaries of technology.

Huawei's European Research Institute (ERI) brings together more than 1,500 researchers across over 20 sites in Europe, including our Research Center in Switzerland.

Our Swiss research laboratory focuses on pioneering next-generation computing architectures and systems for the AI era , spanning innovative hardware, software, and algorithms. Working closely with leading universities and academic partners worldwide, we foster an open, collaborative research environment dedicated to fundamental scientific breakthroughs and long-term innovation.

Our vision is to become a world-leading research center for computing architectures and systems in the AI era, shaping the future of intelligent computing.

About the Team

We are a research team based in Europe, working at the intersection of efficient AI, generative models and distributed systems. We collaborate closely with leading universities, research institutions and engineering teams to make advanced AI workloads practical on modern computing platforms.

Our work includes large multimodal and video-generation systems, but our research questions span a broader range of AI models and workloads. We focus on performance, scalability and efficiency through advances in inference, compilation, distributed execution and AI-driven optimization.

The Role

We are looking for a hands-on Research Scientist with a strong background in AI performance optimization. You will investigate and improve the execution of large AI models and workloads, from kernels and frameworks to end-to-end systems. You will drive research projects within the team’s roadmap: formulating problems, implementing and running experiments, building working systems, measuring improvements and publishing. Experienced candidates may also co-supervise interns and help with recruiting.

What You Will Work On

You will develop and evaluate techniques to improve latency, throughput, memory efficiency and hardware utilization across AI workloads. Example research directions include:

· End-to-end optimization of large-model inference and multi-model workloads, including scheduling, caching and resource allocation

· AI agents for automated performance optimization through profiling, benchmarking, experimentation and iterative refinement

· Compiler, graph and operator optimization, including kernel fusion and low-precision execution

· Distributed inference, communication–compute overlap and GPU/NPU memory management

· Efficient execution of multimodal and generative models, including attention, MoE architectures and video-generation pipelines

· Hardware–software co-design and scalable inference infrastructure for emerging accelerators

Your Responsibilities

· Drive research projects end to end within the team roadmap: from idea to prototype to results at scale

· Implement and optimize model execution, kernels, frameworks or inference systems, and turn results into working systems and demos

· Profile and accelerate models on large NPU/GPU clusters, together with engineering teams

· Take research results to real-time demos and deployable systems

· Define representative benchmarks and quantify latency, throughput, memory efficiency, quality and cost at scale

· Publish at top venues (CVPR, ICCV/ECCV, NeurIPS, ICLR, ICML, SIGGRAPH, MLSYS) and contribute to open-source projects

· Propose new research ideas and directions to the team

· Co-supervise interns and take part in interviews, as a secondary part of the role

What We Are Looking For

· PhD in Computer Science, Electrical Engineering, Artificial Intelligence, Machine Learning or a related field, or equivalent research experience in industry

· Research publications or demonstrated industry research impact in AI systems, efficient ML, compilers or related areas

· Demonstrated hands-on experience optimizing AI models or systems, with measurable improvements in latency, throughput, memory consumption or hardware utilization

· Strong programming skills in Python and experience with at least one relevant ML, compiler or systems programming stack

· Experience profiling, benchmarking and diagnosing performance bottlenecks in ML workloads

· Proven ability to carry research independently, from idea to implementation and result

· Clear communication, in writing and in talks

Experience in one or more of the following areas is an advantage:

· AI-agent-driven optimization, using automated profiling, benchmarking and iterative improvement at one or more levels:

o System/workload: Scheduling, resource allocation, caching and end-to-end latency optimization across models and hardware

o Model/framework: Graph transformations, operator selection, quantization or execution optimization (e.g. PyTorch, TensorRT, vLLM)

o Kernel/operator: Automated kernel generation and tuning (e.g. CUDA, Triton, Ascend C)

· Distributed inference, parallelism, communication–compute overlap and GPU/NPU memory optimization

· Compiler optimization, kernel fusion, graph transformations or low-precision inference (FP8/FP4, MX formats)

· Long-context inference, KV-cache optimization, sparse attention or MoE acceleration

· Step distillation, pruning and optimization of generative models

· Multimodal systems, streaming video generation, VAE decoding or multi-model agent workloads

· Contributions to open-source AI frameworks, compilers or inference systems

What We Offer

· Hands-on research in AI performance optimization and scalable AI systems, with real influence on the team’s direction

· Close collaboration with researchers and engineers from leading universities, research institutions and industry teams

· Access to large-scale AI models and advanced AI computing platforms

· Support for publishing, open-source contributions and conference attendance

· The opportunity to mentor and co-supervise interns

· A highly international research environment in Europe

Who We Are Looking For

We are particularly interested in researchers who can turn promising AI methods into efficient systems on real hardware. Our guiding question is:

· How can we make complex AI models and workloads faster, more scalable and more efficient on real hardware?

If you want to do hands-on research on performance optimization, efficient AI execution and large-scale systems, we would love to hear from you.