Senior CPU Performance Engineer

Arm; Cambridge; On-site; GBP 73,500 - 99,500; Full time

Job summary

Arm is seeking a CPU Performance Engineer in Cambridge to focus on device-level performance analysis. The role involves understanding CPU behaviour using profiling tools and providing insights to guide CPU design decisions. Candidates should have a solid foundation in CPU microarchitecture and experience in performance analysis. The position offers a salary range of £73,500 - £99,500 per year and does not require sponsorship for work in the UK.

Qualifications

• Strong understanding of CPU microarchitecture fundamentals (pipelines, caches, speculation, memory hierarchy).
• Experience in performance analysis (profiling, benchmarking, debugging).
• Familiarity with PMU-based profiling tools or equivalent performance counters.
• Basic programming or scripting skills (Python, C/C++, or similar).
• Strong analytical and problem-solving skills.

Responsibilities

• Analyse performance on real devices using real-world workloads.
• Support investigation of system-level performance issues.
• Use PMU counters and profiling tools to understand CPU behaviour.
• Contribute to identifying performance bottlenecks.

Skills

Strong understanding of CPU microarchitecture fundamentals

Experience in performance analysis

Familiarity with PMU-based profiling tools

Basic programming or scripting skills

Strong analytical and problem-solving skills

Job description

CPU Performance Engineer (Device-Level Analysis)

Overview

We are hiring a CPU Performance Engineer focused on device-level performance analysis. This role sits at the intersection of real-world system behaviour and CPU microarchitecture, with the objective of understanding performance observed on devices and translating it into actionable insights for CPU design teams.

You will work on silicon platforms and real devices, analysing performance across complex and often non-deterministic environments. The role requires a solid foundation in CPU fundamentals, along with a strong interest in understanding how system-level behaviour maps to underlying microarchitecture.

Working alongside experienced architects and performance engineers, you will contribute to investigations, build supporting evidence, and help guide CPU design decisions.

This is not a CPU design or microarchitecture role. Instead, you will influence CPU architecture by identifying bottlenecks, validating hypotheses (where possible) using pre-silicon platforms (e.g. FPGA, emulation, performance models), and driving data-backed recommendations to CPU teams.

Key Responsibilities

Device Performance Analysis

• Analyse performance on real devices using real-world workloads
• Support investigation of system-level performance issues, identifying CPU-related factors
• Operate effectively in low-observability environments (e.g. no waveforms, partial counters, noisy systems)

CPU‑Focused Analysis

• Use PMU counters and profiling tools to understand CPU behaviour
• Contribute to identifying performance bottlenecks (e.g. memory, front‑end/back‑end, execution)
• Build understanding of how microarchitectural features impact real-world performance

Pre-Silicon Support

• Assist with experiments on emulation or FPGA platforms to validate hypotheses
• Reproduce silicon‑observed behaviours in controlled environments to isolate CPU effects
• Where applicable, compare silicon vs model vs emulation to build confidence in findings

Influencing CPU Architecture

• Contribute analysis and data to support discussions with CPU architecture teams
• Highlight gaps between real-world workloads and existing design assumptions
• Document findings and communicate insights clearly to stakeholders

Methodology & Tooling

• Use and progressively improve performance analysis workflows and tools
• Contribute to automation and data analysis where relevant

Required Skills & Experience

• Strong understanding of CPU microarchitecture fundamentals (pipelines, caches, speculation, memory hierarchy)
• Experience in performance analysis (profiling, benchmarking, debugging)
• Familiarity with PMU-based profiling tools or equivalent performance counters
• Basic programming or scripting skills (Python, C/C++, or similar)
• Strong analytical and problem‑solving skills

Preferred Experience

• Exposure to CPU performance analysis or benchmarking
• Familiarity with low-level programming (C / Assembly)
• Basic understanding of RTL or hardware design concepts
• Exposure to FPGA or emulation platforms
• Experience working on real devices (mobile, embedded, or client systems)
• Awareness of system-level performance interactions (OS, scheduling, memory)
• Familiarity with Arm architecture

Candidate Profile

• Interested in understanding how real-world workloads interact with CPU architecture
• Comfortable working with noisy, incomplete, or ambiguous data
• Eager to learn from and collaborate with experienced engineers
• Clear communicator, able to present analysis with appropriate guidance

Additional Information

Please note this role does not meet the eligibility requirements for sponsorship, and therefore the successful candidate must have the right to work in the UK without relying on sponsorship by Arm.

Equal Opportunities at Arm

Arm is an equal opportunity employer, committed to providing an environment of mutual respect where equal opportunities are available to all applicants and colleagues. We are a diverse organization of dedicated and innovative individuals, and don’t discriminate on the basis of race, color, religion, sex, sexual orientation, gender identity, national origin, disability, or status as a protected veteran.

Salary Range

£73,500 - £99,500 per year