Responsibilities
THE ROLE:
AMD is looking for a strategic and hands-on software architect to lead system-wide power and performance optimization for client platforms. You will analyze complete application and operating-system workloads, identify opportunities across hardware and software boundaries, and drive improvements through the AMD software stack and the broader software ecosystem.
In this role, you will work across graphics, display, multimedia, CPU, memory, data-fabric, firmware, drivers, operating systems, and applications. You will develop the system architecture, measurement methodologies, prototypes, and production software necessary to improve performance, energy efficiency, responsiveness, and battery life. You will also collaborate directly with operating-system vendors, OEMs, application developers, and open-source communities to implement sustainable end-to-end solutions, including contributing changes to external codebases when appropriate.
You will be a senior technical leader within a core team of hardware and software specialists working with AMD’s latest technologies and influencing future products.
THE PERSON:
The ideal candidate is a system thinker who can move fluently between workload analysis, software architecture, hands-on implementation, and technical leadership. You are comfortable investigating complex problems that span multiple components and organizations, including areas where ownership or requirements are not initially clear.
You use data and controlled experimentation to separate symptoms from root causes, but you do not stop at analysis. You can define an architecture, build a proof of concept, implement production-quality software, and guide multiple teams through integration and validation.
You communicate complex technical findings clearly, construct persuasive data-driven proposals, and build productive technical relationships across AMD and with external partners. You are willing to challenge existing assumptions and interfaces while remaining pragmatic about product schedules, compatibility, quality, and ecosystem adoption.
KEY RESPONSIBILITIES:
- Define and drive AMD’s system-wide power and performance optimization strategy for key client workloads, including web browsing, productivity, video playback, collaboration, graphics, display, and emerging AI-assisted experiences.
- Serve as a lead software architect across the complete workload path, spanning applications, operating systems, user-mode and kernel-mode drivers, firmware, SoC infrastructure, memory subsystems, and hardware accelerators.
- Analyze real-world workloads and industry benchmarks to identify performance, power, latency, bandwidth, residency, scheduling, and data-movement inefficiencies.
- Develop system-level models and measurement methodologies that connect workload behavior to clocks, voltage states, memory traffic, cache behavior, hardware utilization, and battery-life impact.
- Design and implement software optimizations throughout the AMD stack, including drivers, firmware, libraries, scheduling policies, power-management logic, and hardware-acceleration paths.
- Investigate and modify external and open-source software, including Chromium and relevant Linux components, to improve execution on AMD platforms and enable efficient use of AMD hardware.
- Work directly with operating-system vendors, browser and application developers, OEMs, and ecosystem partners to define requirements, propose API or architectural improvements, contribute patches, and drive solutions through upstream review.
- Develop software proofs of concept to assess new policies, algorithms, APIs, hardware capabilities, and future architectural concepts before product implementation.
- Define AMD reference solutions that balance power efficiency, performance, responsiveness, image quality, compatibility, and scalability across Windows, Linux, and other relevant platforms.
- Build and improve instrumentation, tracing, profiling, telemetry, and visualization capabilities needed to understand temporal system behavior and identify cross-component bottlenecks.
- Establish measurable performance and power baselines, success criteria, regression guardrails, and repeatable test cases for pre-silicon and post-silicon environments.
- Independently validate performance and bandwidth contracts between IP blocks, firmware, SoC infrastructure, and software layers, and drive corrective action when assumptions are not met.
- Lead complex root-cause investigations spanning multiple hardware and software teams, establish clear ownership, and drive issues from initial observation through production resolution.
- Partner with hardware architects to influence future ASIC and SoC designs using workload analysis, competitive data, software experiments, and quantified power/performance trade-offs.
- Evaluate competitive platforms and software pipelines to identify architectural advantages, product gaps, and opportunities for AMD differentiation.
- Provide technical direction and mentorship to engineers, review designs and implementations, and raise the overall system performance and power expertise of the organization.
- Communicate architecture, findings, recommendations, risks, and business impact to engineering teams and senior leadership.
PREFERRED EXPERIENCE:
- Extensive experience in system software, platform architecture, performance engineering, power optimization, or related areas.
- Deep proficiency in C and C++, with demonstrated experience designing, debugging, and optimizing production system software.
- Experience with Windows and/or Linux operating-system architecture, graphics or media stacks, device drivers, firmware, runtime systems, or hardware-accelerated application pipelines.
- Hands-on experience developing and contributing to the Chromium open-source codebase, including the ability to analyze complex browser execution and rendering paths, identify power and performance optimization opportunities, implement changes in Chromium C/C++ code, and work with the upstream Chromium community to review and land production-quality patches.
- Familiarity with open-source development practices, including Git, issue tracking, code review, continuous integration, upstream contribution, and technical engagement with external maintainers.
- Strong understanding of computer architecture, including CPU and GPU execution, memory hierarchies, caching, data fabric, DMA, interrupt handling, scheduling, synchronization, and power-management states.
- Experience analyzing complex workloads using ETW/ETL, performance counters, hardware telemetry, profilers, debuggers, and custom instrumentation.
- Demonstrated ability to correlate application-level behavior with driver, firmware, operating-system, and hardware activity.
- Experience optimizing for multiple constraints, including performance, power, latency, memory bandwidth, quality, reliability, and compatibility.
- Experience developing performance models, conducting controlled experiments, defining KPIs, and translating measurements into architectural or implementation decisions.
- Experience developing proofs of concept and using results to influence software roadmaps or future hardware architecture.
- Ability to navigate ambiguous, cross-organizational problems and drive alignment where responsibility spans several teams or companies.
- Strong technical leadership, communication, and influencing skills, including the ability to present complex findings to both engineering and executive audiences.
- Experience collaborating with operating-system vendors, OEMs, independent software vendors, or open-source communities is highly desirable.
- Experience with graphics, display, multimedia, browser, collaboration, or client power-management workloads is an asset.
- Experience with pre-silicon environments, hardware bring-up, performance verification, or post-silicon optimization is an asset.
ACADEMIC CREDENTIALS:
- Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical field.
- Significant relevant industry experience may be considered in place of an advanced degree.
#LI-TB1
Qualifications
Benefits offered are described: AMD benefits at a glance.
AMD may use Artificial Intelligence to help screen, assess or select applicants for this position. AMD’s “Responsible AI Policy” is available here.
This posting is for an existing vacancy.