ML Infra Network Engineer — Ultra-Fast Interconnects

AI Pulse; Palo Alto (CA); On-site; USD 180,000 - 440,000; Full time

Job summary

SpaceXSI is seeking exceptional ML Infrastructure Engineers to design and optimize the network fabric powering large-scale AI training and inference clusters. You will own end-to-end interconnects from copper to optical, working with vendors and ML teams to ensure scalable, low-latency performance.

The role focuses on 200G+ SerDes, photonics, and system diagnostics, with responsibilities spanning design, validation, and production support in a fast-paced hyperscale environment.

Qualifications

• 8+ years designing, deploying and operating high-speed interconnects.
• Deep knowledge of PAM4 SerDes performance and jitter.
• Experience with optical link budgets and diagnostics.

Responsibilities

• Design, validate, and productize high-speed copper and optical connectivity for AI clusters (100k+ GPUs).
• Own due diligence and onboarding for new 1.6T products including AEC/pluggable transceivers.
• Collaborate with ML teams to translate workload patterns into interconnect topology.
• Perform system-level simulations of fabric performance and drive corrective actions.

Skills

PAM4 SerDes

Optical interconnects

System-level simulation

Vendor collaboration

Thermal/power integrity

Education

Master's or PhD in Electrical Engineering/Photonics/Physics

Job description

SpaceXSI is seeking exceptional ML Infrastructure Engineers to design and optimize the network fabric powering large-scale AI training and inference clusters. You will own end-to-end interconnects from copper to optical, working with vendors and ML teams to ensure scalable, low-latency performance.

The role focuses on 200G+ SerDes, photonics, and system diagnostics, with responsibilities spanning design, validation, and production support in a fast-paced hyperscale environment.