Responsibilities
THE ROLE
Bring AMD's leading-edge SoCs to life in a virtual environment. You'll build Fast Platform Models — high-performance C++ functional models that simulate the software-visible behavior of AMD's next-generation CPUs, MI-series data-center GPUs, Data Fabric and chiplet interconnect, and high-speed IO (USB4, PCIe, IOMMU, RISC-V, and more).
Your models let firmware, OS, driver, and application teams’ boot and debug their software months before silicon exists — and are used by both internal teams and external customers running real workloads.
This is an AI-native role: the leverage comes not from hand-typing C++ but from directing frontier coding models and agents to build, verify, and debug models faster than was possible two years ago. You bring the architectural judgment, spec interpretation, and correctness bar; the AI does the volume. Everyone here works this way — you're either already fluent with AI coding tools or you'll get there fast. You'll partner closely with architecture, firmware, and software teams and learn these systems from the register level up.
If working this way on some of the most complex silicon on the planet is where you want to be, this is your team.
THE PERSON
You think in systems, you love turning dense hardware specs into working models, and you hold a high bar for correctness. You use AI coding tools as a core part of how you work — or you're eager to — and you have thoughts on how to get the most out of them. You're a strong teammate who communicates clearly, thrives with ambiguity, and improves everything you touch.
KEY RESPONSIBILITIES
- Architect and build high-performance C++ functional models of AMD SoCs, IP blocks, and platforms — driving AI coding tools to implement and verify them.
- Enable firmware, OS, driver, and validation teams to develop and debug against your models’ months ahead of silicon.
- Improve the functionality, stability, and performance of existing models, backed by tests that prove correctness.
- Partner with architecture teams to interpret specifications and model next-generation designs.
- Debug internal and customer workloads running on the models.
- Advance our simulation framework, debug tooling, and how the team applies AI across the modeling workflow.
- PREFERRED EXPERIENCE
- Strong C/C++ for high-performance systems software on Linux and/or Windows — able to architect, review, and debug complex code, not just write it.
- Using AI coding assistants/agents as a core part of day-to-day engineering.
- Understanding of hardware/software interfaces (registers, memory maps, interrupts) and subsystem interface protocols.
- Functional modeling, architecture simulation, virtual platform, or hypervisor development — or adjacent experience in pre-silicon verification/emulation, firmware/BIOS bring-up, embedded systems, QEMU/hypervisor, or GPU/driver development.
- x86, ARM, or GPU architecture, drivers, and applications.
- SystemC/TLM-2.0, QEMU, Simics, GEM5, or Virtualizer.
- Linux and/or Windows kernel debugging.
- Multithreaded programming and synchronization primitives (mutexes, atomics, condition variables, lock-free structures), including common pitfalls.
- ACADEMIC CREDENTIALS
- Bachelor's or master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent experience preferred.
- This role is not eligible for visa sponsorship.
Qualifications
