HPC–AI Infrastructure & GPU Cluster Support Engineer

Aptly Technology Corp.; Bengaluru; Hybrid; INR 1,200,000 - 1,800,000; Full time

Job summary

Aptly Technology Corp. in Bengaluru is seeking an experienced HPC–AI Infrastructure Engineer to support large-scale GPU clusters, InfiniBand networking, and data-center operations in a hybrid environment.

You will handle decommissioning, node troubleshooting, incident resolution, and documentation, collaborating with data-center teams and engineering to ensure stability and rapid recovery.

Qualifications

• Experience with HPC, AI infrastructure and GPU clusters.
• Strong knowledge of InfiniBand switches and related networking.
• Hands-on Linux, networking, and infrastructure troubleshooting.
• Experience with monitoring, incident management and change control.

Responsibilities

• HPC–AI Decommissioning: validate cluster readiness and inventory, execute decommissioning per procedures.
• Infrastructure Management: clean up device configuration and coordinate hardware access.
• GPU Node Troubleshooting: diagnose node failures, fault codes, connectivity, and logs.
• Cluster Monitoring: monitor GPU fabric and node health to ensure production stability.
• Incident Management: troubleshoot incidents, coordinate recovery, and document actions.
• Documentation: maintain tickets, change records, and handoff docs.

Skills

HPC infra

GPU clustering

InfiniBand

Linux

Networking

Monitoring

Incident mgmt

Change mgmt

Documentation

Communication

Job description

Job Location: Bangalore/ Hyderabad

Job Mode: Hybrid

Role Summary

Seeking an experienced HPC–AI Infrastructure Engineer to support GPU cluster operations, infrastructure decommissioning, InfiniBand networking, node troubleshooting, and incident resolution in large-scale data center environments.

Key Responsibilities

• HPC–AI Decommissioning: Validate cluster readiness, device inventory, hardware details, and execute approved decommissioning procedures.
• Infrastructure Management: Perform device configuration cleanup, validation, and coordinate with data center teams for hardware access and support.
• GPU Node Troubleshooting: Investigate node failures, fault codes, connectivity issues, and hardware alerts using monitoring tools and system logs.
• Cluster Monitoring: Monitor GPU infrastructure, InfiniBand fabric health, and node availability to ensure production stability.
• Incident Management: Troubleshoot production incidents, coordinate with engineering teams, track recovery activities, and validate service restoration.
• Documentation: Maintain incident tickets, change records, operational trackers, and technical handoff documentation.

Required Skills

• Experience with HPC, AI infrastructure, and GPU clusters.
• Strong knowledge of NVIDIA/Mellanox InfiniBand switches and UFM.
• Hands-on Linux, networking, and infrastructure troubleshooting.
• Understanding of GPU nodes, server hardware, TCP/IP, and data center operations.
• Experience with monitoring, incident management, and change management.
• Strong troubleshooting, communication, and documentation skills.

Preferred Skills

• 1. Experience with large-scale cloud or hyperscale data centers.
• 2. Familiarity with KQL, SQL, Python, Bash, or PowerShell.
• 3. Knowledge of infrastructure automation, GPU node recovery, and hardware lifecycle management.

Good to have :

• CCNA/CCNP
• AZ-900 , AZ-104
• Nvdia networking or infini band related certification (NCA-AIIO)
• Relevant hands on prioritized over certification