We are looking for a hands‑on Technical Lead / Architect to drive PCIe and CXL engineering for AI infrastructure: accelerators, NVMe SSDs, retimers, PCIe switches and AI server platforms. You will own host‑side drivers or device‑side firmware, link bring‑up, debug and validation on customer programs, write and debug code yourself, and lead a team of engineers.
Key Responsibilities
- Develop and debug PCIe/CXL host drivers in the Linux kernel, or device‑side endpoint firmware (bare‑metal or RTOS on ARM/RISC‑V cores).
- Bring up PCIe Gen5/Gen6 links on new platforms: link training (LTSSM), equalization, enumeration, BAR and resource allocation, MSI‑X and DMA.
- Root‑cause link, error (AER/DPC), hot‑plug and performance issues across hardware, firmware and OS layers.
- Build and run PCIe/CXL validation: compliance, interoperability, stress, error injection and telemetry, automated in Python.
- Integrate accelerators, NVMe SSDs, SmartNICs/DPUs, PCIe switches and retimers into AI server platforms.
- Work directly with customer silicon, hardware and validation teams; lead design reviews and technical customer discussions.
- Own technical delivery for the team and mentor engineers.
Qualifications
Education
- Bachelor’s or master’s degree in computer engineering or computer science, Electronics, Electrical, (or equivalent experience).
Experience
- 8 to 14 years in embedded systems software.
- At least 4 years of hands‑on PCIe work on a shipped product or silicon program.
Must-Have Technical Skills
- Strong C programming, with hands‑on coding.
- Either Linux kernel PCIe driver development or firmware development for a PCIe device/endpoint.
- Lab debug with PCIe protocol analyzers/exercisers, JTAG, and kernel or firmware traces.
- Python for test automation.
Good-to-Have Technical Skills
- CXL 2.0/3.x (CXL.io, CXL.mem), memory expansion and pooling.
- PCIe Gen6 (PAM4, FLIT mode), retimers and PCIe switches.
- PCI-SIG or CXL compliance testing.
- GPU or custom AI accelerator integration.