Staff Engineer Accelerated Compute (GPU Infrastructure)

As a Senior / Staff Engineer in Accelerated Compute (GPU Infrastructure) at a rapidly growing AI infrastructure startup based in Bengaluru, your main role is to own the GPU infrastructure powering a new AI-focused cloud. You will be responsible for designing and automating systems to simplify high-performance computing through API calls. By joining early, you will have the opportunity to influence architecture and define how enterprise customers manage large AI workloads. **Key Responsibilities:** - Own end-to-end GPU enablement across the platform, from design to production support. - Implement and manage NVIDIA vGPU across multiple hypervisors for efficient GPU sharing among several VMs. - Extend Kubernetes to be GPU-aware and develop custom device plugins for MIG instances. - Utilize NVIDIA MIG to partition GPUs into hardware-isolated slices that can be rented as units. - Create and maintain pre-configured VM images optimized for GPU workloads, including necessary drivers and libraries like MPI and NCCL. - Develop automation allowing customers to deploy multi-node, InfiniBand-connected GPU clusters with a single API call. **Qualifications Required:** - Proven background in cloud, SRE, or infrastructure engineering with a focus on compute. - Hands-on expertise with NVIDIA GPU technologies such as vGPU, MIG, and CUDA. - Strong knowledge of hypervisors like KVM, VMware, or XenServer. - Experience with high-performance networking technologies such as InfiniBand, RDMA, and IPoIB. - Proficiency in production-grade Kubernetes skills. This position offers you the chance to work with cutting-edge technologies in AI infrastructure and cloud computing. Joining this startup will allow you to contribute to shaping the future of AI-first cloud infrastructure.,