Job Description
Join our customer's team working with an elite customer, developing cutting-edge AI training, inference server infrastructure silicon. As a Network Stack Developer, you will be developing performant L2, L3 Protocol stack for validating multiple AI silicon under development.
Key Responsibilities
- Develop, validate and maintain high-speed L2, L3 protocol stacks over NIC drivers in Linux kernel, user mode and other host, device operating systems.
- Implement and optimize L2, L3 protocols, RDMA, and RoCE for high-performance networking applications.
- Collaborate with cross-functional teams to develop, test NIC firmware and protocol stack software, troubleshoot and resolve NIC-related technical issues.
- Develop and maintain test frameworks for NIC performance and functionality testing.
Required Skills
- Proficiency in C, C++ and Python programming languages.
- In-depth knowledge of networking protocols, including TCP, IP, IPv4 and IPv6 and their packet formats.
- Hands-on experience with tools for capturing and analyzing protocol packet exchanges across high-speed interfaces.
- Understanding of high-performance multi-core CPU, GPU clusters and their high-speed interconnect fabric interfaces such as PCIe and Infiniband.
- Working familiarity with Host Channel Adapters (HCAs) or bridges that parse, translate and route data from standard NW stacks into fabric-specific transport logic would be an added advantage.
Desirable Skills
- Python scripting for test log analysis and NIC behavior triage.
- Experience with machine learning and deep learning frameworks and their underlying NW payload transfer.
- Working experience with latest AI coding copilot-based flows to generate, review, test and release code.
- Hands on experience in developing AI training, inference application would be valuable.
Education
Bachelor's or Master's degree in Computer Science, Electrical Engineering or a related field with a focus on computer systems or networking.
Experience
At least 6+ years of experience in L2, L3 stack development, using modern NIC device driver APIs including kernel mode and user mode drivers. Experience working with high speed cloud data center server stack development.
Work Location: In person
