AI Infra Engineer - GPU Fleet & Automation

The Consensus; Turkey; On-site; TRY 400,000 - 800,000; Full time

Benefits offered by this job

Interesting and challenging work

Learning and growth opportunities

Regular team events and offsites

Job summary

fal is an on-site role in Turkey for a hands-on engineer tasked with maintaining a large fleet of GPU servers. You will build software and processes to keep the fleet healthy and productive, including provisioning, health monitoring, and error detection.

You will drive recovery with partners when automation falls short, and you will implement robust security and performance tuning to support AI workloads.

Qualifications

• 3+ years managing large server fleets (100+ nodes) at scale.
• Strong Python production tooling experience.
• Deep Linux knowledge including kernel tuning and I/O.
• Experience with infrastructure-as-code (Ansible, Terraform).
• Storage systems like NVMe, NFS, Lustre/GPFS.

Responsibilities

• Build and maintain Python fleet tracking and provisioning.
• Automate provisioning, health checks, and GPU diagnostics.
• Create dashboards and alerts for hardware health.
• Develop automated error detection and recovery processes.
• Collaborate with partners to resolve technical issues.

Skills

Python programming

Linux systems

Infrastructure as code

GPU infrastructure

Performance tuning

Automation

Communication

Self-starter

Tools

NVIDIA drivers

CUDA

PXE/iPXE provisioning

libvirt/KVM

GPFS/Lustre

NFS

Job description

fal is an on-site role in Turkey for a hands-on engineer tasked with maintaining a large fleet of GPU servers. You will build software and processes to keep the fleet healthy and productive, including provisioning, health monitoring, and error detection.

You will drive recovery with partners when automation falls short, and you will implement robust security and performance tuning to support AI workloads.