Director of Backend Engineering, AI Infrastructure

About the role

Coupang Intelligent Cloud is looking for a visionary Director of Backend Engineering to own the software platform that powers its global AI training and inference infrastructure. You will lead the design of SDN orchestrators, fleet health automation, and high‑performance storage back‑ends that abstract complex GPU hardware into a reliable cloud service.

Key responsibilities

• Lead the design and delivery of an SDN orchestrator to automate GPU networking (InfiniBand, RoCE, NVLink) and data‑center routing.
• Oversee backend services for GPU health monitoring, fault detection, and automated RMA workflows.
• Drive capacity planning, traffic engineering, load balancing (NGINX/Kong) and IP address management for zero‑bottleneck AI training.
• Collaborate with storage engineers to build interfaces for parallel file systems (Lustre, Weka, VAST) and AI object storage with high durability and low latency.
• Act as the final technical authority for AI Infra architecture, ensuring multi‑region resilience and sub‑millisecond coordination.
• Champion a “Hardware‑as‑Code” culture using Python, Ansible and Terraform, and establish 24/7 operational standards for >99.99% fleet availability.

Required profile

• 15+ years of backend engineering experience, including at least 5 years in a senior leadership role.
• Deep knowledge of Linux internals, hardware‑software interfaces, and distributed systems.
• Strong expertise in L2/L3 networking and specialized fabrics such as InfiniBand or RoCE.
• Proven experience with high‑performance storage systems (GPFS, Lustre, parallel file systems).
• Understanding of GPU architecture, power envelopes, liquid‑cooling constraints, and NVIDIA/HPE/Dell hardware.

Required skills

• Python, Go, C++
• Terraform, Kubernetes, Ansible
• SDN, InfiniBand, RoCE, NVLink
• NGINX, Kong, IPAM
• Lustre, GPFS, Weka, VAST
• NVIDIA GPU, GPUDirect technologies