N

Nio

Automotive-grade SoC designer for autonomous driving

Senior Staff Performance Tuning Engineer - CPU PMU & Virtualization

Systems Software EngineeringConfirmed live in the last 24 hours
Full-timeSan Jose, CA, USA

About the Job

Requirements

  • Bachelor’s, Master’s, or Ph.D. in Computer Science, Electrical Engineering, or a related field.
  • At least 8 years of experience in system performance engineering or related roles.
  • Strong understanding of computer architecture, including CPU pipelines, execution engines, memory hierarchy, interrupt architectures such as GICv3/GICv4 and APIC concepts, MMU, virtual memory, and page table structures.
  • Deep expertise with Linux performance tooling, including perf, eBPF, BCC, bpftrace, ftrace, and trace-cmd.
  • Proficiency in C/C++, Python, and shell scripting for performance tooling and automation.

Responsibilities

  • Identify and analyze performance bottlenecks across CPU, memory hierarchy, cache, interconnect, and interrupt subsystems.
  • Use hardware PMU counters to measure and interpret cycles, instructions, cache misses, branch mispredictions, TLB misses, and stall cycles.
  • Compute and analyze IPC, CPI, MPKI, memory bandwidth utilization, and stall breakdowns.
  • Optimize system software, applications, and drivers for throughput, latency, and performance determinism.
  • Profile and reduce hypervisor overhead, including VM exits, interrupt injection, and stage-2 MMU translation costs.
  • Optimize vCPU placement strategies, huge page usage, and IRQ affinity for performance-sensitive workloads.
  • Analyze and tune virtio and IOMMU performance paths.
  • Investigate cross-core and cross-cluster latency issues, including IPI overhead and interrupt routing on ARM-based systems.
  • Configure and interpret ARM PMUv3 and related vendor-specific performance monitoring extensions.
  • Apply Top-Down Microarchitecture Analysis (TMA) to identify frontend and backend bottlenecks.
  • Correlate PMU-derived metrics with kernel traces, source code paths, and system behavior.
  • Design and implement custom event sets tailored to workload-specific performance characterization.

Desired Qualifications

  • Hands-on experience with ARM PMUv3 on Cortex-A55/A76/A78 or Neoverse platforms.
  • Experience with one or more virtualization stacks, including KVM/QEMU, Xen, seL4, or other RTOS/hypervisor environments.

About the Company

Nio

Automotive-grade SoC designer for autonomous driving

GeniTech designs and develops high-performance automotive system-on-chips for autonomous driving. Its NX9031 SoC uses a 5nm automotive process with 50+ billion transistors, includes a high-dynamic-range ISP and a heterogeneous many-core pool, and is built with a full-stack in-house approach to tightly synchronize sensors and compute cores for real-time fusion and decision-making. It differentiates itself by combining internal hardware supply to its parent company with external IP licensing and joint ventures, offering both an integrated platform and licensed technology. The goal is to industrialize automotive semiconductors and monetize proprietary compute platforms for autonomous driving through licensing, royalties, and collaborations in automotive and robotics.

Company size
5,001-10,000
Company stage
Early VC
Total funding
$310.3M
Headquarters
Huangpu, China
Founded
2014

Recently Posted Jobs

Explore more opportunities at Nio

See more jobs