Openai
SoC Engineer
San Francisco, Remote. Posted a month ago
The role
You will architect, design, and integrate first-party AI accelerators. This involves building processors, interconnects, I/O, and platform infrastructure to create complete functional systems.
What you will do
- Define and implement SoC architectures spanning embedded processors, AMBA interconnects, peripheral subsystems, PCIe, Ethernet and die-to-die interfaces.
- Develop microarchitecture and RTL design of subsystems supporting I/O and SoC functionality.
- Integrate Arm Cortex-M and Cortex-A processors, AXI/APB fabrics, and third-party controller IP.
- Develop reset and boot architectures working with firmware and software teams.
- Incorporate debug, test, and security requirements into the SoC architecture and implementation.
- Drive PCIe and other I/O integration including controller and PHY interfaces, high-speed SerDes, and die-to-die connectivity.
- Contribute to power management and security architecture and design.
- Work with third-party IP vendors to define requirements, evaluate capabilities, and resolve integration issues.
- Work across architecture, design, verification, physical design, and software teams.
- Build and apply automation and AI-assisted tools to accelerate design and debugging.
Required qualifications
- Experience in SoC architecture, RTL design, and subsystem or full-chip integration.
- Experience with hands-on ownership through implementation and silicon bring-up.
- Familiarity with Arm processor subsystems, including Cortex-M and Cortex-A.
- Familiarity with AMBA protocols and interconnects such as AXI and APB.
- Understanding of PCIe protocols and experience configuring, integrating, and debugging controller IP.
- Experience integrating high-speed I/O, including SerDes/PHY interfaces and die-to-die links.
- Experience working directly with third-party IP vendors.
- Understanding of reset and boot architecture and hardware, firmware, and software interactions.
- Experience designing for debug, test, and security.
- Comfortable writing and debugging Verilog/SystemVerilog.
- Ability to build scripts or tools to make integration more efficient and reliable.
About the company
OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. Its Hardware organization develops custom silicon and systems for advanced AI workloads.