Apply on Employer Site

Advanced Microdevices Pvt. Ltd. (India) · 1 month ago

Senior GPU Firmware Engineer

Santa Clara, CA

Full-time

Hybrid

Senior Level

Advanced Micro Devices, Inc is a company focused on building products that accelerate next-generation computing experiences. The Senior GPU Firmware Engineer will join the Datacenter firmware application team to support GPU customers, collaborating with partners and internal teams to optimize AMD's Instinct™ Accelerators for deployment in data centers.

BiopharmaBiotechnologyIndustrialManufacturing

Responsibilities

Manage technical interaction with OEM/ODM Partners to enable deployment of AMD Instinct™ Accelerators in Partner systems

Work alongside hardware and upper software layers to co-optimize the whole AI software stack

Design and build tools for better collecting/presenting GPU performance details corelating to low-level hardware characteristics

Support Partners in the bring-up and validation of AMD Instinct™ GPUs in their system, guide partners on use of AMD tools, qualification test methods, and analysis of test results

Lead the debug of Partner/Customer issues (firmware, HW, driver), working with a cross-functional team and driving the root cause investigation

Work with Partners on the development of manufacturing/screen tests to ensure reliability at scale

Understand Partner requirements and schedule, identify gaps in AMD offering and work with other stakeholders to close them

Author design guideline, technical presentations, and training material

Provide recommendation to improve customer experience with our SW and HW

Qualification

Firmware developmentEmbedded software developmentC/C++ programmingSystem level debuggingPower managementOS kernel internalsGPU architecturesEnterprise LinuxAnalytical skillsCommunication skillsProblem solvingCollaboration

Required

Familiar with embedded/firmware development

Familiar with GPU driver/runtime

Familiar with OS kernel internals

Familiar with microcontroller fundamentals

Familiar with hardware power/frequency controls

Comfortable performing quantitative analysis of workload

Hands-on technical problem solver

Ability to collaborate directly with customers and internal engineering teams

Strong analytical instincts and a structured approach to problem solving

Clear and proactive communication across technical and non-technical audiences

Ability to collaborate across hardware, firmware, and software disciplines

Curiosity, creativity, and persistence in solving complex engineering challenges

Preferred

Experience with firmware developments

Experience with embedded software development

Experience with power management and control theory

Experience working on system level reliability and resiliency features

Familiarity with OS kernel/driver internals

Familiarity with GPU architectures and runtimes

Familiarity with microcontroller fundamentals (caches, buses, memory controllers, DMA, etc.)

Strong C/C++ programming skills

Strong knowledge in PC/server architecture and interfaces

Experience with system level debug

Strong System Level debugging skills with hands-on experiences in system bring-up, HW debug, and performance optimizations on various system architectures

Understanding and experience working with Enterprise Linux environment (Ubuntu, CentOS/RHEL and SLES)

Excellent oral and written communication skills to communicate technical results clearly and accurately

Experience or knowledge of server firmware/BIOS settings, boot process, server monitoring and management SW

Solid knowledge of Shell/BASH, C/C++, Python, or other framework

Experience with OpenCL, CUDA, or ROCm is a plus

BS/MS (Computer Science, Computer Engineering, Electrical Engineering, or related equivalent)