Perception Systems Performance Lead for Hardware Architecture

Summary

In this position, you will lead efforts within Platform Architecture team to characterize in-system performance of perception systems in Rivian vehicles, with the aim to drive hardware architecture decision making.  This is a hybrid software / hardware role that requires a software background to identify, bring-up, develop and measure key system-level performance metrics, and the necessary embedded hardware background to support these activities. You will collaborate cross-functionally with hardware design and software application teams to accomplish this work, and thereby drive direction of next generation hardware architectures, as well as motivate optimization of current generation platforms.  This role focuses on the characterization of perception related hardware systems, ranging from the sensors and data acquisition pipeline, through to the computational performance of the respective application processors.

Responsibilities

• Lead the hands-on efforts to characterize perception related hardware across various vehicle systems (e.g. surround view camera systems, security monitoring systems, ADAS, and others), ranging from sensor and data pipeline performance, through to downstream application computation characterization.
• Collaborate with hardware architecture, hardware design and software application teams to create comprehensive characterization plans to surface key performance metrics, locate bottlenecks and chokepoints, identify over-/under-design scenarios relative to application requirements, etc.
• Execute characterization plans, develop analytics, and present clear validation metrics and associated reports to help architecture planning and decision-making
• Build test equipment, benchtop mock-ups, prototypes, and/or POCs as necessary to evaluate sub-system performance, and/or investigate characterization of future-facing hardware options.
• Bring-up and develop software on evaluation kits, early prototypes boards, vehicle system mock-ups, to facilitate execution of characterization plans
• Contribute to the evaluation and design review of next generation architectures, leveraging insights from performance characterization efforts

Qualifications

Required Competencies

• Advanced degree in Computer Science, Electrical Engineering or a related field, and a track record of hands-on experience in relevant industries, internships and/or academic settings
• Direct background with camera sensors, radars and/or downstream processing pipelines, including their data formats, ISPs, DSPs, SerDes, and how to characterize and validate their respective key performance metrics.
• Strong software and hardware understanding of embedded computing architectures, and related characterization methods for quantifying system-level performance of their processors (e.g. CPUs, GPUs, ML/DL and other accelerator cores), memory systems (e.g. RAM, caches, and storage), networking / peripheral interfaces, and serialization pipelines.
• Embedded perception platforms coding experience related to some combination of sensor/pipeline bring-up, system software or applications development, system-level h/w performance characterization, testing automation / scripting / instrumentation for software performance evaluation
• Conversant with software stacks, drivers and APIs for embedded perception computing architectures typical in automotive settings, including GPUs, ML/DL and other accelerators, and respective performance characterization methodologies.
• Innate curiosity to figure out how systems work by decomposing them through the application of first-principles, an ability to translate this into measurable metrics, and a drive to optimize design performance by these means

Optional Cross-Functional Skillsets

• Understanding of automotive network architectures and protocols, and how to characterize and validate relevant performance metrics at the component and system levels (e.g., throughput, latency, collisions, input-to-actuation delays, photons-to-photons latencies, etc.)
• Grasp of technologies in embedded display systems, ranging from panel technologies to protocols and PHYs (e.g. DSI, eDP, etc.), through to display pipelines and drivers in embedded SoCs, and the myriad of related performance characterization tools and benchmarks.

Desirable Skillsets

• Familiarity with automotive and/or autonomous vehicle requirements
• Conversant with automotive protocols (e.g., CAN, LIN, CAN-FD, Ethernet, FPDLINK, etc.)
• Experience with contract manufacturing and related supply chains
• Experience working on high volume products in aggressive design environments