Principal Solution Architect - Passage

We are now looking for an outstanding Principal Solution Architect to join our team. The role’s primary responsibility is to work collaboratively with customers in architecting the next generation of super computers and machine learning systems enabled by Lightmatter’s photonic interconnect technology Passage.

In this role, you will work closely with Tier 1 Semiconductor companies and Cloud Service Provider organizations that are defining the next generation of scale out, HPC, and machine learning systems. You will be working cross functionally within Lightmatter as you define an optimal technical solution. Additionally, you will be collaborating with customers who are domain experts in HPC and ML system architectures to refine these designs based on their application requirements. You should be comfortable supporting deep technical discussions focused on interconnect topologies and how to scale HPC/ML architectures in 1024+ node systems. Excellent interpersonal skills and the ability to work independently are essential to being successful in the role.

Responsibilities

• Partner with Lightmatter sales leaders to develop strong relationships with key technical stakeholders, and help them optimize their system architecture by taking advantage of Lightmatter’s interconnect technology
• Be a domain expert in high performance computing (HPC) and machine learning system architectures with a focus on high-performance interconnects
• Leverage background in HPC/scientific computing algorithms to identify performance bottlenecks related to interconnect and optimize topologies to improve performance
• Apply extensive knowledge in high-performance interconnect topologies and function as an expert and principal advisor
• Define new and refine existing ideas and design concepts that address scaling HPC and ML system architectures using Lightmatter’s highly disruptive interconnect technology Passage
• Explains advanced concepts in HPC/ML system architecture and understands how those concepts relate to other disciplines.

Requirements

• BS or MS in Engineering, Mathematics, Physics, or Computer Science (or equivalent experience).
• 10+ years of work-related experience in a high-tech electronics industry in a system design role or technical customer support role.
• Experience defining the system architectures of high-performance compute clusters focused on HPC and/or ML workloads
• Customer facing skill set and background as well as the ability to communicate effectively with customers
• Action-oriented with strong analytical and problem-solving skills
• Strong time-management and organization skills for coordinating and prioritizing multiple projects and initiatives
• Excellent written and oral communications skills with the ability to effectively interface with management and engineering
• Strong team-working and interpersonal skills

Preferred Qualifications

• C/C++ and Python programming skills

• Master's or PhD in related science or engineering field

• High Performance Computing & Networking
• Experience architecting systems with computational accelerators and high-performance interconnects (e.g., NVLink, Infiniband, etc...)
• Proficiency profiling scientific computing, and/or machine learning workloads to identify performance bottlenecks related to networking, storage, and compute.