About The RoleOur client's Embedded Architecture team, you will work across CPU, GPU, Real-Time Unit (RTU), Video, and Physical AI technologies, helping shape the future of heterogeneous computing platforms. You will join a core architecture group at the forefront of innovation, driving technology strategy and influencing next-generation silicon, software, and platform roadmaps. This role provides a unique opportunity to analyze and optimize complex AI, vision, perception, planning, and control workloads across heterogeneous compute engines including CPUs, DSPs, FPGA fabric, GPUs, and AI accelerators, while helping define the architecture of future intelligent systems and programmable computing platforms.
This role will help shape our client's Physical AI strategy and future programmable computing platforms!
Key Responsibilities- Define sensor-to-actuator Physical AI reference architectures.
- Analyze and partition AI, vision, perception, planning, and control workloads across CPU, DSP, FPGA fabric, GPU, and AI accelerators.
- Drive architecture tradeoffs involving performance, latency, power, memory bandwidth, and determinism.
- Develop workload models, benchmarks, and performance projections for robotics and autonomous systems.
- Collaborate with customers, ecosystem partners, and internal silicon teams to identify future platform requirements.
- Influence next-generation SoC, FPGA, memory, interconnect, and accelerator roadmaps.
QualificationsMinimum Qualifications:- PhD in Electrical Engineering, Computer Engineering, Computer Science, Robotics, or related field.
- 1+ years of experience and research interest in system architecture, embedded computing, AI acceleration, robotics, or heterogeneous computing.
- 1+ years understanding of CPU, DSP, FPGA, GPU, AI accelerator, memory, and interconnect architectures.
- Experience with AI/ML, computer vision, robotics, or autonomous systems workloads.
- Proven track record of technical leadership, innovation, and architecture definition.
Preferred Qualifications- Experience with ROS2, edge AI, machine vision, autonomous systems, or industrial robotics.
- Knowledge of FPGA-based acceleration and heterogeneous computing platforms.
- Publications, patents, or recognized contributions in AI, robotics, or computer architecture.
