Global Infoways

Location: Client sites across India & HQ in Noida Full-time Early-career (2 4 yrs) About Global Infoventures For over two decades, Global Infoventures (GI Ventures) has been a trusted technology partner to Indias education sector, building our foundation on comprehensive ERP solutions. Today, as a 5-Star NVIDIA Partner, we have evolved to become a leading integrator of AI supercomputing infrastructure. We design, deploy and operate NVIDIA powered AI Centres of Excellence, empowering universities to tackle national-scale challenges in scientific research and AI. Why this role matters You will be one of the engineers who keep multi-crore AI clusters running flawlessly. In just 12 months, you will learn to build, tune and support NVIDIA DGX platforms skills that place you in the top percentile of HPC talent nationwide. Key outcomes (first 12 months) Deploy and validate new DGX nodes and Ethernet/InfiniBand fabrics within agreed go- live windows Maintain 99% infrastructure uptime through proactive monitoring and root-cause analysis Resolve service tickets within SLA by collaborating with GI Ventures/NVIDIA support and network teams Write and continuously improve SOPs, shell scripts, and Kubernetes manifests Conduct health checks and firmware upgrades with minimal unplanned downtime Train site administrators on basic DGX, Linux and Kubernetes workflows Core responsibilities Installation & Configuration Rack, cable and image DGX servers using NVIDIA BCM; configure BMC/IPMI, RAID and high-speed switches. System Administration Manage Linux (Ubuntu/DGX OS) environments, user quotas, security patches and GPU drivers. Container & Orchestration Operate Docker/Containerd images and assist with Kubernetes cluster lifecycle. Performance & Reliability Profile CPU/GPU, NIC and storage metrics; recommend optimisations. Customer Support Serve as on-site technical contact; document incidents and deliver post-mortems. Minimum qualifications: 1 2 years in Linux or network support with hands-on CLI troubleshooting Proficiency with shell scripting Working knowledge of TCP/IP, subnets, VLANs, SSH and basic switch configuration Familiarity with containers and fundamental Kubernetes objects (Pods, Nodes, Services) Ability to explain GPU vs CPU roles and use BMC/iDRAC/IPMI for remote administration Proficiency with Git for version-controlled configs and documentation Great-to-have extras: Built a Linux server from bare metal Deployed a small-scale K8s cluster Exposure to network switch configuration and storage arrays What you ll gain: Elite hardware access Daily work on DGX clusters rarely available outside hyperscalers Structured mentorship Direct coaching from our AI team and certified DGX architects Visible impact Your fixes keep high-value research projects online and productive Transparent rewards Competitive salary and transparent rewards based on cluster uptime and client feedback