Lenovo

We are looking for a AI Network Architect for the GPU cluster solutions we develop for the Cloud Service Providers market.

Key Responsibilities:

• Design network architectures for datacenter AI infrastructure solutions, covering different network sizes and types such as IB, Ethernet, scale-up, scale-out, and so on.   
• Design and implement GPU cluster reference designs with specific network topologies, switch/NIC configurations, rack layouts, BOMs, etc.
• Design and implement methodologies, processes, software frameworks and tools for network performance evaluation and tuning in GPU clusters.
• Research network technology trends, collaborate with ecosystem partners, and participate in industry efforts such as Ultra Ethernet, UALink, and so on.
• Collaborate with cross-functional teams and contribute to business projects and customer engagements.

• BS/MS in Computer Science, Computer Engineering, Electrical Engineering, or other related fields.
• 8+ years of experience in datacenter network design and performance tuning.
• Strong knowledge about network protocols (e.g., Ethernet, Infiniband).
• Strong programming skills (e.g., C/C++, Python, Go, shell scripting).
• Excellent communication and leadership skills.
• Ability to learn new technologies and concepts quickly.
• Innovative mindset with the ability to think outside the box.

Preferred Qualifications:

• PhD in Computer Science or Computer Engineering.
• Experience with GPU rack-scale and cluster networking in production environments.
• Experience with networking software and hardware (NOS, switch, NIC, etc.).
• Experience with network performance evaluation and tuning.
• Knowledge about AI workload characteristics and software stack such as collective communications.
• Ability to innovate and publish at top conferences.