Power Attainment Engineer

THE ROLE:

The successful candidate will assume responsibility for silicon and system level electrical validation, maximizing performance/watt through power management feature tuning and optimization, power model correlation and prototyping activities of AMD Datacenter GPU products.

You will interact with key System and Power Management Architects, Firmware teams, Power Modelling Leads, Board Design & Validation Engineers, Performance engineers, Customer Engineering teams as well as Product Definition team to achieve the desired product goals.

THE PERSON:

The ideal person is passionate about technology and building great products.

You have strong Power fundamentals, understanding of Power Management features, hands on experience optimizing them, silicon validation experience, good understanding of PID controllers, PDN, Physical design, Semiconductor Process, Thermal/Power interactions, Power/Performance optimization & Power Models background.

You have excellent communication skills, critical problem-solving skills, data analysis and visualization skills, able to multi-task and work with cross-domain, cross-functional teams to build state of the art HPC & AI products.

Must be a self-starter, strong team collaborator and able to independently drive tasks to completion working with cross-functional teams.

You are hands-on, willing to interact directly with the hardware, utilize scopes and probes to gather detailed electrical information.

You have automation, scripting, data processing and analysis skills.

Have leadership and mentorship skills to lead, train and groom college grads or junior engineers in the team.

KEY RESPONSIBILITIES:

• Execute Power Attainment test plans in post-silicon phases in support of Data Center GPU product roadmap optimizing for power, perf/watt and performance.


• Configure and setup ML/AI Datacenter GPU systems for data collection, experiments and test plan execution


• Utilize lab equipment such as oscilloscope, high speed probes, function generator and data acquisition equipment to gather required electrical characterization data for power and performance optimization.


• Actively participate in analysis of post silicon performance and power data collected to ensure integrity of results, provide summary and conclusions of results, drive productization of features


• Analyze and debug interactions between various power management features


• Analyzing data from workload or execution output datalogs using excel or JMP analysis tools manually or developed automation


• Execute ROI analysis of power management features and provide feedback to power management architecture team.


• Support prototyping experiments and development of new GPU features that impact performance and power


• Electrically stress the system, validate the limits of ASIC and system/board components and optimize settings for stability and performance.


• Troubleshoot system-level issues that may occur in test environments and platforms


• Proactively drive continuous improvement for post-silicon power attainment activities


• Participate in development of automation environment in developing scripts automating workloads, enhancing capabilities of execution capabilities in Linux, Python and other support software support tools


• Work in a fast-paced resource constrained environment to build top of the line HPC & AI GPU products

PREFERRED EXPERIENCE:

• 5+ years of hands-on experience as an engineer in semiconductor industry.


• Demonstrated ability to execute and deliver multiple projects in a timely fashion.


• Prioritizing work items in a fast-paced environment and escalating as necessary.


• Excellent grasp of computer organization/architecture, GPU architecture and power management


• Knowledge in power limited performance methodologies and control theory


• Extensive experience in platform optimization.
  
  Solid knowledge of Computer I/O.


• Experience with tools for power and performance analysis


• Strong programming skills, scripting experience in Python preferred


• Familiarity with HPC/AI applications, benchmarks would be a big plus.


• Desirable to be proficient in Linux command line environment and Shell scripting


• Deep knowledge of power management techniques like deep sleep, clock gating, pstates etc


• Experience with container technologies (ex.
  
  Docker)


• Strong analytical and problem-solving skills with a key attention to detail


• Experience in data analysis, summarization, and presentation


• Excellent presentation and communication skills


• Experience in use and debug of lab tools such as oscilloscopes, DAQs, power measurement capabilities

ACADEMIC CREDENTIALS:

• Bachelors or Masters degree in electrical or computer engineering or computer science.
  
  MS Preferred.

#LI-RL1

Benefits offered are described: AMD benefits at a glance.

AMD does not accept unsolicited resumes from headhunters, recruitment agencies, or fee-based recruitment services.

AMD and its subsidiaries are equal opportunity, inclusive employers and will consider all applicants without regard to age, ancestry, color, marital status, medical condition, mental or physical disability, national origin, race, religion, political and/or third-party affiliation, sex, pregnancy, sexual orientation, gender identity, military or veteran status, or any other characteristic protected by law.

We encourage applications from all qualified candidates and will accommodate applicants' needs under the respective laws throughout all stages of the recruitment and selection process.