Benchmark AI inference workloads using PyTorch, ONNX Runtime, vLLM, llama.cpp, TensorFlow, and related frameworks and runtimes.
Analyze key performance metrics, including latency, throughput, time to first token, tokens per second, memory utilization, power consumption, and performance per watt.
Profile workloads and identify bottlenecks across AI frameworks, runtimes, libraries, drivers, operating systems, firmware, memory subsystems, and hardware.
Lead complex root-cause analysis and recommend optimizations related to quantization, graph execution, kernels, memory usage, workload scheduling, and heterogeneous compute utilization.
Conduct competitive benchmarking across AMD and competing CPU, GPU, APU, and AI accelerator platforms.
Collaborate with SoC architects, framework, runtime, compiler, driver, firmware, and system software teams to improve AI workload performance.
Develop and improve automation for benchmark execution, data collection, regression analysis, and results visualization.
Translate benchmarking data into actionable technical recommendations that influence software optimization, hardware design, product strategy, and customer engagements.
Present technical findings and recommendations to engineering teams, customers, executives, and other cross-functional stakeholders.
Provide technical leadership, mentor engineers, and drive alignment across multiple teams.
Required Skills & Experience
10 to 14+ years of relevant experience in AI performance engineering, system performance analysis, workload optimization, benchmarking, or post-silicon performance validation.
Hands-on experience evaluating, benchmarking, and characterizing AI inference workloads on CPU, GPU, APU, NPU, or heterogeneous computing platforms.
Strong experience with PyTorch, ONNX Runtime, vLLM, llama.cpp, TensorFlow, or equivalent AI frameworks and inference runtimes.
Deep expertise in platform architecture, hardware performance, and system-level optimization, including CPU, GPU, NPU, memory, power, and thermal subsystems.
Strong Linux systems expertise, including performance profiling, workload tracing, memory management, workload scheduling, automation, and system-level debugging.
Experience with profiling tools such as Linux perf, AMD uProf, ROCm Profiler, Nsight Systems, VTune, Perfetto, or equivalent tools.
Strong understanding of hardware architecture, memory hierarchy, data movement, power and thermal behavior, and performance trade-offs.
Experience with AI optimization techniques such as quantization, mixed-precision inference, graph optimization, operator fusion, kernel tuning, and memory optimization.
Proficiency in Python and Bash for benchmark automation, data analysis, and reporting.
Working knowledge of C or C++ for performance debugging and low-level workload analysis.
Ability to independently drive benchmarking strategy, competitive analysis, root-cause investigation, and technical recommendations.
Excellent executive, customer-facing, and cross-functional communication skills.
Ability to convert complex performance data into clear technical reports, dashboards, presentations, and actionable recommendations.
Proven ability to provide technical leadership and drive complex initiatives across multiple engineering teams.
Preferred Qualifications
Experience with ROCm, Ryzen AI Software, Vitis AI, MIGraphX, or related AI technologies.
Experience with MLPerf Inference or comparable AI benchmarking methodologies.
Knowledge of INT8, INT4, FP16, BF16, and mixed-precision execution.
Experience with large language model inference, including batching, KV-cache management, model parallelism, and memory optimization.
Experience with Docker, virtualization, and containerized AI environments.
Exposure to post-silicon bring-up, early silicon testing, competitive benchmarking, or customer-facing performance analysis.
Experience influencing platform architecture, software roadmaps, product strategy, or customer solutions through data-driven performance analysis.
Education
Bachelor’s or master’s degree in Computer Engineering, Electrical Engineering, Computer Science, Artificial Intelligence, Machine Learning, or a related technical field.
#LI-PK1
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.
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.
Lead AI Power & Performance Engineer
Role Overview
We are seeking an experienced AI Power & Performance Engineer, Lead to join our AI System Power and Performance team. The candidate will lead the evaluation, benchmarking, profiling, and optimization of AI inference workloads across modern CPU, GPU, APU, NPU, and heterogeneous computing platforms.
This senior technical role requires deep expertise in AI workload performance, platform architecture, and system-level optimization. The candidate will independently drive benchmarking strategy, complex root-cause analysis, competitive evaluation, and technical recommendations while collaborating with engineering, product, customer, and executive stakeholders.
Key Responsibilities
Define and drive the AI performance benchmarking and workload characterization strategy across CPU, GPU, APU, NPU, and heterogeneous platforms.
Benchmark AI inference workloads using PyTorch, ONNX Runtime, vLLM, llama.cpp, TensorFlow, and related frameworks and runtimes.
Analyze key performance metrics, including latency, throughput, time to first token, tokens per second, memory utilization, power consumption, and performance per watt.
Profile workloads and identify bottlenecks across AI frameworks, runtimes, libraries, drivers, operating systems, firmware, memory subsystems, and hardware.
Lead complex root-cause analysis and recommend optimizations related to quantization, graph execution, kernels, memory usage, workload scheduling, and heterogeneous compute utilization.
Conduct competitive benchmarking across AMD and competing CPU, GPU, APU, and AI accelerator platforms.
Collaborate with SoC architects, framework, runtime, compiler, driver, firmware, and system software teams to improve AI workload performance.
Develop and improve automation for benchmark execution, data collection, regression analysis, and results visualization.
Translate benchmarking data into actionable technical recommendations that influence software optimization, hardware design, product strategy, and customer engagements.
Present technical findings and recommendations to engineering teams, customers, executives, and other cross-functional stakeholders.
Provide technical leadership, mentor engineers, and drive alignment across multiple teams.
Required Skills & Experience
10 to 14+ years of relevant experience in AI performance engineering, system performance analysis, workload optimization, benchmarking, or post-silicon performance validation.
Hands-on experience evaluating, benchmarking, and characterizing AI inference workloads on CPU, GPU, APU, NPU, or heterogeneous computing platforms.
Strong experience with PyTorch, ONNX Runtime, vLLM, llama.cpp, TensorFlow, or equivalent AI frameworks and inference runtimes.
Deep expertise in platform architecture, hardware performance, and system-level optimization, including CPU, GPU, NPU, memory, power, and thermal subsystems.
Strong Linux systems expertise, including performance profiling, workload tracing, memory management, workload scheduling, automation, and system-level debugging.
Experience with profiling tools such as Linux perf, AMD uProf, ROCm Profiler, Nsight Systems, VTune, Perfetto, or equivalent tools.
Strong understanding of hardware architecture, memory hierarchy, data movement, power and thermal behavior, and performance trade-offs.
Experience with AI optimization techniques such as quantization, mixed-precision inference, graph optimization, operator fusion, kernel tuning, and memory optimization.
Proficiency in Python and Bash for benchmark automation, data analysis, and reporting.
Working knowledge of C or C++ for performance debugging and low-level workload analysis.
Ability to independently drive benchmarking strategy, competitive analysis, root-cause investigation, and technical recommendations.
Excellent executive, customer-facing, and cross-functional communication skills.
Ability to convert complex performance data into clear technical reports, dashboards, presentations, and actionable recommendations.
Proven ability to provide technical leadership and drive complex initiatives across multiple engineering teams.
Preferred Qualifications
Experience with ROCm, Ryzen AI Software, Vitis AI, MIGraphX, or related AI technologies.
Experience with MLPerf Inference or comparable AI benchmarking methodologies.
Knowledge of INT8, INT4, FP16, BF16, and mixed-precision execution.
Experience with large language model inference, including batching, KV-cache management, model parallelism, and memory optimization.
Experience with Docker, virtualization, and containerized AI environments.
Exposure to post-silicon bring-up, early silicon testing, competitive benchmarking, or customer-facing performance analysis.
Experience influencing platform architecture, software roadmaps, product strategy, or customer solutions through data-driven performance analysis.
Education
Bachelor’s or master’s degree in Computer Engineering, Electrical Engineering, Computer Science, Artificial Intelligence, Machine Learning, or a related technical field.
#LI-PK1
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.
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.
