Description

You will design and develop system-level profiling tools for CPU, memory, IO, and power analysis on AMD platforms.

This role is on-site.

Responsibilities

• Build and optimize data collection frameworks leveraging hardware counters, IBS, and OS tracing mechanisms.
• Develop low-overhead profiling infrastructure for large-scale workloads to ensure minimal performance impact.
• Enhance performance analysis pipelines including data processing, correlation, and visualization for actionable insights.
• Enable cross-platform profiling support across Linux, Windows, and FreeBSD.
• Integrate support for runtime and framework-level tracing including OpenMP, MPI, Java, and Python.

Required Skills

• 10+ years of experience with C++ and System Design.
• Strong understanding of microarchitecture, pipelines, caches, and memory hierarchy.
• Hands-on experience with hardware performance counters and performance analysis.
• Deep knowledge of Operating Systems internals and Linux.
• Experience with cross-platform development and system-level debugging.
• Familiarity with profiling tools and techniques.

Preferred Skills

• Experience with AMD uProf, perf, VTune, or rocProfiler.
• Knowledge of distributed/HPC workloads (MPI, OpenMP) and trace analysis.
• Experience with AI/ML-assisted performance insights.

Key Skills

Education

Any Graduate