S
Senior AI Accelerator System Architect
В офисеПолная занятостьAI EngineeringSemiconductor Manufacturing
Join a research and engineering team developing next-generation deep learning accelerator platforms. Own system-level performance optimization and hardware-software co-design from algorithmic concepts through silicon implementation.
Обязанности
- Define accelerator architecture, including compute units, memory hierarchies, and dataflow, for machine learning inference workloads
- Lead performance analysis using roofline models and analytical profiling
- Drive design-space exploration and multi-objective optimization under power, performance, and area constraints
- Lead hardware-software co-design and specify hardware features and dataflow patterns that improve compiler efficiency
- Collaborate with algorithm researchers to map large language models and generative AI workloads to hardware
- Develop pre-silicon performance models to validate architecture decisions
Требования
- At least 7 years of experience in processor or AI accelerator architecture, high-performance computing, or performance engineering
- Expertise in roofline analysis, workload characterization, and design-space exploration
- Experience with agentic workflows and autonomous optimization systems
- Strong understanding of computer architecture, memory systems, and parallel processing
- Bachelor’s degree in Computer Engineering, Computer Science, or Electrical Engineering
Будет плюсом
- M.Sc. or Ph.D. in a relevant field
- Experience with Black Box Optimization and Hyperparameter Optimization
- Track record in hardware-software co-design for AI accelerators
- Experience with compiler architecture or MLIR
- Publications or patents in computer architecture or machine learning systems