Qambria

The role

We are seeking a highly skilled Senior HPC Performance Engineer to join our team. This role focuses on analyzing, optimizing, and scaling system performance across next-generation HPC and Quantum platforms. You will work at the intersection of hardware and software, driving efficiency and throughput by leveraging high-speed interconnects and advanced memory management techniques.

What you'll do

Performance Engineering: Profile, benchmark, and optimize memory management and high speed interconnect for HPC and Quantum workloads.
Interconnect Optimization: Design, analyze, and tune performance of high-speed interconnects including PCIe, NVLink, InfiniBand, and RoCE, within nodes and across distributed systems.
Memory & Data Movement: Develop and optimize DMA and RDMA workflows, ensuring low-latency, high-bandwidth data transfer across GPUs, CPUs, NICs, and control hardware.
System-Level Analysis: Investigate bottlenecks across compute, memory, and networking layers; propose architectural improvements and software optimizations.
Collaboration: Work closely with hardware architects, driver developers, and application teams to co-design solutions that maximize performance.
Benchmarking: Define and execute performance benchmarks, scaling studies, and proof-of-concept experiments to validate system performance at scale.
Innovation: Stay up-to-date on emerging interconnect technologies, memory hierarchies, and distributed workload management strategies.

‍

What we're looking for

Required:

PhD in Computer Science, Physics, Mathematics, Electrical/Computer Engineering or related field — or equivalent industry experience.
Strong background in HPC, distributed systems, or high-performance networking.
Expertise in one or more interconnect technologies: PCIe Gen4/Gen5, NVLink, InfiniBand, RoCE.
Experience with DMA, RDMA, and low-latency data transfer mechanisms.
Familiarity with Linux kernel, drivers, and system-level performance tuning.
Hands-on experience with HPC benchmarks and profiling tools (e.g., MPI profilers, Nsight Systems, perf).
Proficiency in C/C++ and/or low-level systems programming; scripting in Python or Bash a plus.

Preferred:

Experience in hybrid quantum-classical algorithms/applications.
Experience porting and running applications for multi-CPU, multi-GPU and FPGA systems.

‍

Thank you!
Your submission has been received!

Oops!
Something went wrong while submitting the form.

Senior HPC Performance Engineer

Switzerland

Full-time

The role

What you'll do

What we're looking for