Senior Software Engineer - CPU Performance Modeling

AMD - Advanced Micro Devices

Iasi, Romania

THE ROLE:

As a Core Performance Modelling Engineer, you will join AMD's Core Performance team to help model, analyze, and improve the performance of next-generation CPU cores. You will work with performance simulators, correlation regressions across simulation, emulation, and RTL, and internal analysis tooling. You will be digging into complex microarchitectural behavior to root-cause performance discrepancies, and you will collaborate closely with teammates to validate and improve performance models. This role also offers the opportunity to apply machine learning and AI-driven techniques to large-scale performance datasets, working alongside colleagues already developing automated analysis and triage tooling for the team.

KEY RESPONSIBILITIES:

Develop, debug, and maintain software and scripts supporting CPU core performance modelling and correlation regressions across simulation, emulation, and RTL environments;
Investigate and root-cause discrepancies in performance metrics (e.g., IPC, PMC data, latency) between simulation, emulation, and RTL results, collaborating with design and verification teams;
Analyze processor pipeline and memory hierarchy behavior to help identify the source of performance and correlation issues;
Contribute to internal performance analysis dashboards and tooling, including data pipelines and automated data processing/validation;
Apply machine learning and AI techniques to performance datasets, supporting automated analysis, triage, and root-cause tooling in collaboration with teammates driving these efforts;
Write clean, maintainable, well-documented code and clearly communicate technical findings to the team;
Participate in regular team syncs and cross-team coordination with design, verification, and emulation teams.

PREFERRED EXPERIENCE:

Strong programming and algorithmic problem-solving skills; solid computer science fundamentals;
Proficiency in C++ and scripting languages (Python preferred);
Solid foundational knowledge of modern computer architecture, including processor pipelines and memory hierarchies;
Experience with Unix/Linux development environments and version control tools;
Familiarity with hardware verification concepts and techniques (e.g., simulation, debug, coverage) is a plus;
Hands-on academic, project-level, or internship experience with logic design or Hardware Description Languages (Verilog, SystemVerilog, or VHDL) is a plus;
Familiarity with, or interest in, machine learning/AI techniques and applying them to engineering data is a plus;
Knowledge of assembly language is a plus but not required;
Strong debugging skills across multi-component systems;
Excellent written and spoken English, with the ability to communicate technical concepts clearly.