Track Lead (Support & Operations) (IN)
HCLTech
HCLTech
Seeking a skilled Track Lead for Support & Operations. This role requires deep expertise in data center networking, particularly with BGP and VXLAN-EVPN overlays. A strong understanding of GPU cluster networking, Linux, and Kubernetes is essential. You will play a key role in designing, deploying, and troubleshooting complex network environments.
Familiarity with NVIDIA's networking hardware and software, including Spectrum-X and DPUs, is highly desirable. You'll also need to understand HPC environments and the nuances of rail-optimized architectures for efficient traffic flow.
The ideal candidate possesses excellent troubleshooting skills, down to the optics and cabling level, and can quickly adapt to new technologies like Cumulus Linux. Experience with AI training and inference concepts will be beneficial.
Architect, deploy, troubleshoot, and optimize large-scale production clusters with a focus on NVIDIA DGX and GPU administration.
Develop advanced solutions for congestion control, traffic engineering, topology design, and network performance optimization.
Leverage extensive experience in automated provisioning using shell scripting, Python, Ansible, Salt, APIs, and CI/CD pipelines.
Ensure seamless integration between networking, GPU computing, distributed storage, and HPC workloads.
Create High-Level Design (HLD) and Low-Level Design (LLD) documents, validating all designs prior to production rollout.
Demonstrate deep hands-on experience with multiple network operating systems, including Cumulus Linux, SONiC, IOS, Junos OS, and EOS.
Proven hands-on experience with BGP (eBGP underlay, BGP unnumbered, ECMP) in leaf-spine/Clos data center fabrics.
Expertise in the design and deployment of VXLAN-EVPN overlays, IRB, and multi-tenancy with VRFs.
Solid understanding of Linux and the ability to rapidly learn Cumulus Linux.
Strong hands-on experience with Kubernetes and Linux.
Good working knowledge of HPC environments.
Familiarity with NVIDIA Spectrum-X Ethernet platform, ConnectX SuperNICs, and BlueField DPUs.
Basic conceptual knowledge of InfiniBand, RDMA/RoCEv2, and lossless Ethernet tuning (PFC, ECN, DCQCN).
Understanding of GPU cluster network design based on DGX reference architectures and rail-optimized architectures.
Proficiency in fabric monitoring and troubleshooting using streaming telemetry and Linux networking tools.
HCLTech
IT Consulting