SMTS Platform Emulation Engg

• AMD is seeking a Platform Emulation Engineer to join our Data Center GPU organization
• Our products support the rapidly scaling Data Center and High-Performance Compute infrastructure
• You will be an integral member of the Platform emulation team responsible for emulation infrastructure development, execution and debug of critical firmware used in our HPC and ML products, developing test plans and execute test cases to enhance silicon test coverage, train and support validation teams on emulation technology, build and execute industry standard HPC benchmarks and applications on the emulator
• As a platform emulation engineer you will have an opportunity to work on bleeding edge SoC architecture and technology and participate in development of emulation infrastructure to enable pre-silicon activities to ensure high quality Silicon and fast time to market
• You will work alongside our team of innovative engineers to support the deployment of AMD s Instinct ML products targeting Supercomputers and Data Center workloads

KEY RESPONSIBILITIES

• CPU/GPU Firmware/OS bring up and debug using leading edge emulators (Veloce, Zebu, Palladium)
• Develop/modify/enhance functional/stress tests to find silicon bugs on emulator
• Validate features in firmware including but not limited to RAS, Resets, Power management, Secure boot, Virtualization
• Validate multi-socket configurations in emulation using firmware
• Bring up the full AMD graphics software stack in emulation
• Bring up ML/AI applications in emulation Collaborate with workloads and SW teams, run apps in multi-socket configurations, collect and interpret results for debug and tuning
• Checkout of IP features in software simulation model as a pre-step for emulation
• Interface with Central Emulation team responsible for helping to develop, debug and establish CPU/GPU Emulation environment
• Provide debug leadership, input to test plans and hands on execution
• Work with extended teams to debug environment, drive cross team testing (Firmware, Software, Diags/tools, Validation, Apps)
• Triage failures with design, verification, firmware, software, and emulation teams
• Research and enable new tools and infrastructure to support emulation activities
• Run and collect data for analysis on AMD s high-end emulators and simulation models
• Develop scripts/tools to parse data from emulation runs
• Instrument firmware code to support failure debug
• Run and collect functional and performance data for HPC workloads
• Attend weekly meetings, status communication, presentations.

• Solid experience working on emulation platforms such as Veloce, Zebu, or Palladium (compilation, debug, performance, and throughput tuning)
• Debugging experience is a must particularly in HW/SW co-debug environments
• Ability to trace code across multiple domains and root cause using waveforms and other tracing tools
• Understanding of Verilog, VHDL design
• Understanding of System Verilog, UVM verification environments is a plus
• Knowledge of computer hardware architecture (Graphic card, CPU/APU, memory, bus logic, and display technologies)
• Knowledge of memories (LPDDR5, HBM3/4, SPI, etc) and Bus arch (PCIe, CXL, AMBA, DFI, etc)
  Knowledge of computer software architecture and boot flow (bootrom, firmware, bios, device drivers, OS)
• Knowledge of secure platform concepts such as Root of Trust, Chain of Trust, FW signing, encryption
• Knowledge of platform management HW and SW components such as BMC, SMC and board management software libraries
• Knowledge of HPC, Parallel programming (OpenMP, OpenMPI), heterogeneous platforms consisting of CPU and GPU a plus
• Working knowledge of Linux/Unix environment and shell scripting
• Experience in Building and Installing software packages on Linux platform a plus
• Programming skills, C, C++, Python, Tcl, Ruby, assembly (ARM & x86)
• Familiarity with bug tracking tools such as Jira
• Excellent oral and written communication skills