Senior AI Compute Engineer

Neysa; Mumbai; On-site; INR 900,000 - 1,500,000; Full time

Job summary

neo cloud is seeking a Senior AI Compute Engineer to design, deploy, and operate GPU clusters (NVIDIA & AMD) for LLM training and HPC workloads. You will own end-to-end provisioning from architecture to production rollout and provide expert Linux optimization and Kubernetes orchestration.

You will manage GPU infrastructure, CUDA, drivers, NVLink, DGX-inspired setups, Slurm and MPI, and implement infrastructure-as-code workflows with Python, Ansible, and Terraform, ensuring reliability and

Qualifications

• 8+ years of Linux systems administration and data center deployment.
• 3+ years HPC infra experience with Slurm/PBS and parallel computing.
• Expertise in Linux admin (RHEL, Ubuntu, Rocky) including kernel tuning and driver management.
• Proficiency with GPU infra (NVIDIA GPUs, CUDA, NVLink, GPUDirect RDMA).
• Experience with Kubernetes and container orchestration (Helm, Docker, Containerd).

Responsibilities

• Deploy and manage AI GPU clusters for enterprise and cloud customers from planning to production acceptance.
• Manage Linux systems with kernel tuning and performance optimization at scale.
• Build and optimize GPU infra: configure CUDA, NVIDIA drivers, NVLink, NVSwitch.
• Deploy and operate Kubernetes clusters with GPU support using Helm, Docker, and Containerd.
• Configure and optimize Slurm, MPI, and parallel file systems for AI training and HPC workloads.
• Perform root cause analysis on production incidents and reduce cluster issues through validation and monitoring.

Skills

Linux systems administration

HPC infrastructure

GPU compute infrastructure

Kubernetes & containers

Python scripting

Education

Bachelor's degree in CS/EE/IT

Tools

CUDA

NVIDIA drivers

Docker

Kubernetes

Terraform

Git

Job description

We are building next-generation AI infrastructure powering LLM training, inference clusters, and HPC workloads. As a Senior AI Compute Engineer, you will design, deploy, and operate GPU clusters based on NVIDIA and AMD GPU infrastructure—managing everything from hardware configuration and Linux optimization to Kubernetes orchestration and customer success. You will work with team end-to-end: from architecture planning and production rollout through optimization and technical support. This is hands-on infrastructure engineering at neo cloud.

About the Role

We are building next-generation AI infrastructure powering LLM training, inference clusters, and HPC workloads. As a Senior AI Compute Engineer, you will design, deploy, and operate GPU clusters based on NVIDIA and AMD GPU infrastructure—managing everything from hardware configuration and Linux optimization to Kubernetes orchestration and customer success. You will work with team end-to-end: from architecture planning and production rollout through optimization and technical support. This is hands-on infrastructure engineering at neo cloud.

What you will be doing:

• Deploy and manage AI GPU clusters (NVIDIA and AMD) for enterprise and cloud customers—end-to-end ownership from planning to production acceptance
• Manage advanced Linux systems (RHEL, Ubuntu, Rocky) with expertise in kernel tuning, driver optimization, and system performance at scale
• Build and optimize GPU infrastructure: configure CUDA, NVIDIA drivers, GPU Operator, GPUDirect RDMA, NVLink, and NVSwitch
• Deploy and operate Kubernetes clusters with GPU support using Helm, Docker, and Containerd for AI workload orchestration
• Configure and optimize Slurm, MPI, and parallel file systems for distributed AI training and HPC workloads
• Perform root cause analysis on production incidents and proactively reduce cluster issues through validation and monitoring

What we need to see: Core Compute (8+ years)

• 8+ years of hands-on Linux systems administration and data center infrastructure deployment
• 3+ years of HPC infrastructure experience with job schedulers (Slurm/PBS) and parallel computing
• Expertise in Linux administration (RHEL, Ubuntu, Rocky)—kernel tuning, driver management, PCIe troubleshooting, performance optimization
• Proficiency with GPU infrastructure (NVIDIA GPUs, CUDA, GPUDirect RDMA, NVLink, DCGM monitoring and troubleshooting)
• Experience with Kubernetes and container orchestration (Helm, Docker, Containerd, GPU Operator, CSI drivers)

Automation & Infrastructure-as-Code

• 3+ years of infrastructure automation using Python, Bash, Ansible, Terraform, or SaltStack
• Ability to develop provisioning workflows, CI/CD pipelines, and version control with Git
• Strong scripting skills to automate deployment, validation, and operational tasks at scale

Ways to stand out from the rest:

• NVIDIA certifications (AI Infrastructure, AI Operations, Certified Associate/Professional)
• Kubernetes certifications (CKA, CKS) or Red Hat Certified Engineer (RHCE)
• Experience with AI Factory deployments, LLM training clusters, or GPU cloud platforms
• Background with NVIDIA DGX SuperPOD, HGX clusters, or NVIDIA Spectrum-X networking
• Experience with monitoring stacks (Prometheus, Grafana, DCGM, ELK, Loki) and observability in distributed systems
• Hands-on experience with advanced storage systems (Ceph, GPFS, Weka, VAST) or bare-metal provisioning (MAAS, Foreman)

Minimum Qualifications:

• Bachelor's degree in Computer Science, Electrical Engineering, Electronics, Information Technology, or equivalent professional experience
• 8+ years of Linux systems administration and data center deployment
• 4+ years of consulting or customer-success engineering roles

Soft Skills:

• Strong problem-solving and debugging abilities across hardware, kernel, and application layers
• Ownership mindset with accountability for deployment quality and customer success
• Cross-functional collaboration with other teams
• Proactive approach to continuous learning and staying current with AI infrastructure trends
• Strong documentation and presentation skills, able to defend design decisions amongst peers.