Networking SME - AI Infrastructure

Hamilton Barnes Associates Limited; United States; On-site; USD 225,000 - 275,000; Full time

Benefits offered by this job

Full benefits

Architectural ownership

Autonomy

Job summary

Hamilton Barnes Associates Limited seeks a Networking SME to design and implement high-performance networking across Apple Silicon nodes for enterprise-scale AI workloads. You will enable multiple chips to function coherently as a single system and own NCCL integration with the network layer.

Work closely with kernels and runtime engineers to ensure tight full-stack integration. The role offers autonomy on a small, highly technical team and a base salary of $250,000 with full benefits.

Qualifications

• Hands-on experience with NVLink and high-bandwidth chip-to-chip interconnects.
• Strong background in RoCE and high-performance Ethernet fabric design.
• Experience with InfiniBand networking in HPC or large-scale AI workloads.
• Proficiency with NCCL and understanding of how collective communication interacts with the network.

Responsibilities

• Design and implement high-performance networking across Apple Silicon nodes.
• Develop interconnect strategies to operate multiple chips as a single logical computer.
• Own NCCL-layer integration to enable efficient collective communication.
• Define scalable networking architecture for the platform.
• Collaborate with kernels and runtime engineers for full-stack integration.

Skills

NVLink interconnects

RoCE

InfiniBand networking

NCCL

Apple Silicon architecture

Job description

Looking for a role with plenty of growth opportunities?

Join a company delivering managed Apple Silicon infrastructure for enterprise-scale agentic AI workloads. With more than 10,000 Apple Silicon instances deployed across multiple configurations, the platform provides managed infrastructure with full system access, consistent performance, reliable operations, and enterprise-grade SLAs.

The organization is seeking a Networking SME to help build dedicated Apple Silicon infrastructure for next-generation AI workloads. This role focuses on solving complex networking challenges to enable multiple discrete chips to operate as a single, unified system.

Responsibilities:

• Design and implement high-performance networking across Apple Silicon nodes, drawing on technologies like NVLink, InfiniBand, and RoCE
• Develop interconnect strategies that allow multiple chips to operate coherently as a single logical computer
• Own the NCCL-layer integration, ensuring collective communication operates efficiently across our hardware
• Define the networking architecture that underpins the entire platform and scales with our infrastructure
• Work closely with kernels and runtime engineers to ensure tight integration across the full stack

Skills/Must Have:

• Hands-on experience with NVLink, designing or operating high-bandwidth chip-to-chip interconnects
• Strong background in RoCE and high-performance Ethernet fabric design
• Experience with InfiniBand networking in HPC or large-scale AI training environments
• Proficiency with NCCL and an understanding of how collective communication libraries interact with the network layer
• Nice to have: experience with Apple Silicon architecture

Benefits:

• Full Benefits
• Own the networking architecture of a platform being built from the ground up, making foundational technical decisions.
• Work on an AI infrastructure challenge at the forefront of the industry that few engineers have tackled.
• Join a small, highly technical team with real autonomy and direct impact.

Salary:

• $250,000 Base Salary