We are seeking an AI + HPC Hardware Performance Researcher to explore and advance software–hardware co-design innovations for next-generation artificial intelligence and high-performance computing (HPC) systems.
You will join a research-driven compute architecture team focused on large-scale, heterogeneous computing systems. The team investigates workload characterization, runtime management, performance modeling, architecture simulation, and prototyping to achieve orders-of-magnitude improvements in performance, energy efficiency, and total cost of ownership.
In this role, you will work across the full computing stack—from applications and system software to microarchitecture and semiconductor technologies. You will collaborate closely with experts from multiple disciplines to identify impactful co-design opportunities and to build the modeling and evaluation infrastructure needed to assess them. This position plays a key role in shaping the future of high-performance computing architectures.
Key Responsibilities
You will:
* Research innovative solutions for next-generation large-scale HPC clusters, spanning workload characterization, resource management, system architecture, and microarchitecture
* Develop performance, power, cost, and emissions models for future hardware at various levels of abstraction to identify bottlenecks and propose improvements
* Capture and analyze AI workload characteristics within HPC applications as inputs to hardware and system-level models
* Analyze performance on existing systems to validate and calibrate models
* Collaborate across teams to guide HPC hardware/software co-design decisions, based on key hardware limitations and software performance constraints
* Model architectural features and system configurations that improve performance, energy efficiency, total cost of ownership, and environmental impact
Required Knowledge and Skills
* Master’s or PhD degree in Computer Science, Computer Engineering, or a related STEM field (early-career experience is welcome)
* Experience with performance analysis of HPC applications, such as molecular dynamics or computational fluid dynamics
* Experience with AI within HPC workloads is a strong plus
* Experience with performance modeling and/or computer architecture simulation across multiple hardware platforms (e.g. CPU, GPU, NPU, or interconnects)
* Solid understanding of heterogeneous system architectures, including memory systems and microarchitecture
* Good understanding of machine learning techniques and their performance implications
* Strong programming skills in C++ and Python; CUDA experience is a plus
* Strong communication and collaboration skills, enabling effective work in a dynamic, distributed, and multidisciplinary team
* Excellent English communication skills, as you will work in an international and multicultural environment