Senior Staff System Engineer, GPU Fleet

Coupang · Bengaluru · 12+ yrs experience · Posted 2026-07-18

Tech stack: Linux, Python

Apply on the company site · Get a referral for this role

Coupang salary & ratings · Coupang interview process · More live openings

About the role

Responsibilities:
- We are seeking a Sr Staff System Engineer, GPU Fleet for our Coupang Intelligent Cloud (CIC) team, to serve as the senior technical owner for our hyperscale GPU compute infrastructure.
- In this role, you will define fleet architecture, drive reliability and automation at scale, and lead the operation and evolution of GPU systems supporting large‑scale AI training and inference workloads.
- This is a hands‑on, staff‑level individual contributor role with broad technical ownership, high operational impact, and significant cross‑functional influence across hardware, infrastructure, and datacenter operations.
- CIC builds the infrastructure for abundant intelligence.
- We partner with leading AI labs, governments, and enterprises to deliver hyperscale GPU compute with high reliability, performance, and efficiency.
- Our infrastructure supports some of the most demanding AI training and inference workloads in production today.
- We operate with urgency, deep ownership, and a strong bias toward execution.
- Reliability, operational excellence, and rigorous systems engineering are core to our business.
- As a Sr Staff System Engineer, GPU Fleet, you will be the senior technical owner for CIC’s large‑scale GPU compute infrastructure.
- This is a hands‑on senior individual contributor role with fleet‑level responsibility and broad cross‑functional influence.
- You will define the technical direction for how GPU fleets are architected, operated, automated, and evolved across multiple generations of hardware.
- Your work will directly affect fleet reliability, operating efficiency, scalability, and customer success.
- This role does not involve people management, but it carries principal‑level scope, autonomy, and decision‑making authority across infrastructure, hardware, and operations.
- Fleet Architecture & Technical Ownership
- Own the end‑to‑end technical architecture of hyperscale GPU fleets, including hardware platform selection, firmware strategy, OS configuration, drivers, networking, and observability.
- Define and enforce technical standards and best practices for fleet reliability, availability, performance, and operability.
- Lead major fleet‑wide initiatives such as new GPU platform bring‑ups, multi‑generation hardware transitions, and architectural redesigns.
- Evaluate trade‑offs across cost, performance, reliability, and time‑to‑deploy, and make technically sound decisions under ambiguity.
- Reliability, Availability & Performance
- Set and drive fleet‑level reliability, availability, and performance objectives.
- Lead root‑cause analysis and resolution of complex, systemic failures affecting large portions of the fleet or multiple datacenters.
- Identify recurring failure patterns and drive long‑term fixes spanning hardware, software, automation, and operational processes.
- Work directly with hardware vendors and partners to resolve platform‑level issues and influence future hardware designs.
- Automation & Systems Engineering
- Design and build large‑scale automation systems for:
- GPU fleet provisioning and lifecycle management
- GPU health validation, diagnostics, and certification
- Automated remediation, recovery, and replacement workflows
- Eliminate manual operational toil through durable, well‑designed tooling that scales with fleet growth.
- Ensure all fleet systems are observable, testable, and resilient under failure conditions.
- Operational Leadership
- Act as a senior escalation point for critical production incidents impacting GPU availability or customer workloads.
- Participate in on‑call rotations with a strong emphasis on preventing future incidents, not just responding to them.
- Lead high‑severity post‑incident reviews and ensure learnings are translated into concrete engineering and process improvements.
- Technical Influence & Mentorship
- Provide technical mentorship and guidance to system and infrastructure engineers across the organization.
- Serve as a trusted technical partner to platform engineering, networking, datacenter operations, and leadership teams.
- Influence CIC’s long‑term infrastructure roadmap through strong technical judgment and data‑driven recommendations.
Qualifications:
- 12+ Years of overall experience with at least 8+ years of experience in Linux systems engineering, infrastructure engineering, or datacenter operations, operating production environments with strict uptime and performance requirements.
- Deep, hands‑on expertise in Linux system internals, including process scheduling, memory management, filesystem behavior, networking, kernel behavior, and system performance analysis.
- Demonstrated experience operating hardware‑intensive infrastructure in production, including bare‑metal servers at scale.
- Proven ability to debug complex issues across multiple system layers, including hardware components, firmware/BIOS, kernel drivers, OS configuration, and user‑space services.
- Extensive experience writing production‑grade automation using Python and Bash for provisioning, configuration management, diagnostics, remediation, and fleet operations.
- Strong understanding of how to design systems that are observable, resilient, and safe under failure, rather than reliant on manual intervention.
- Direct experience operating large‑scale GPU fleets supporting AI/ML training and/or inference workloads in production.
- Familiarity with modern GPU platforms and ecosystems, including GPU drivers, CUDA, NCCL, and high‑performance compute workloads.
- Experience with high‑speed interconnects and datacenter networking, such as NVLink, InfiniBand, RDMA, and high‑throughput Ethernet.
- Prior ownership of fleet‑wide or platform‑wide initiatives, such as new hardware bring‑ups, major architectural changes, or reliability transformations.
- Experience partnering directly with hardware vendors or manufacturers to troubleshoot systemic issues or influence future platform designs.
- Strong intuition for failure modes at scale, including cascading failures, correlated faults, and second‑order effects across systems.
- History of acting as a technical authority or escalation point for ambiguous, high‑impact production problems.
- Ability to mentor engineers through design reviews, technical problem solving, and modelling strong operational ownership.
- Experience participating in on‑call rotations and responding to high‑severity production incidents with clear ownership, urgency, and technical leadership.
- Strong written and verbal communication skills, including clear post‑incident reviews and technical documentation.
- Hybrid

Qualifications

- 12+ Years of overall experience with at least 8+ years of experience in Linux systems engineering, infrastructure engineering, or datacenter operations, operating production environments with strict uptime and performance requirements.
- Deep, hands‑on expertise in Linux system internals, including process scheduling, memory management, filesystem behavior, networking, kernel behavior, and system performance analysis.
- Demonstrated experience operating hardware‑intensive infrastructure in production, including bare‑metal servers at scale.
- Proven ability to debug complex issues across multiple system layers, including hardware components, firmware/BIOS, kernel drivers, OS configuration, and user‑space services.
- Extensive experience writing production‑grade automation using Python and Bash for provisioning, configuration management, diagnostics, remediation, and fleet operations.
- Strong understanding of how to design systems that are observable, resilient, and safe under failure, rather than reliant on manual intervention.
- Direct experience operating large‑scale GPU fleets supporting AI/ML training and/or inference workloads in production.
- Familiarity with modern GPU platforms and ecosystems, including GPU drivers, CUDA, NCCL, and high‑performance compute workloads.
- Experience with high‑speed interconnects and datacenter networking, such as NVLink, InfiniBand, RDMA, and high‑throughput Ethernet.
- Prior ownership of fleet‑wide or platform‑wide initiatives, such as new hardware bring‑ups, major architectural changes, or reliability transformations.
- Experience partnering directly with hardware vendors or manufacturers to troubleshoot systemic issues or influence future platform designs.
- Strong intuition for failure modes at scale, including cascading failures, correlated faults, and second‑order effects across systems.
- History of acting as a technical authority or escalation point for ambiguous, high‑impact production problems.
- Ability to mentor engineers through design reviews, technical problem solving, and modelling strong operational ownership.
- Experience participating in on‑call rotations and responding to high‑severity production incidents with clear ownership, urgency, and technical leadership.
- Strong written and verbal communication skills, including clear post‑incident reviews and technical documentation.
- Hybrid

Responsibilities

- We are seeking a Sr Staff System Engineer, GPU Fleet for our Coupang Intelligent Cloud (CIC) team, to serve as the senior technical owner for our hyperscale GPU compute infrastructure.
- In this role, you will define fleet architecture, drive reliability and automation at scale, and lead the operation and evolution of GPU systems supporting large‑scale AI training and inference workloads.
- This is a hands‑on, staff‑level individual contributor role with broad technical ownership, high operational impact, and significant cross‑functional influence across hardware, infrastructure, and datacenter operations.
- CIC builds the infrastructure for abundant intelligence.
- We partner with leading AI labs, governments, and enterprises to deliver hyperscale GPU compute with high reliability, performance, and efficiency.
- Our infrastructure supports some of the most demanding AI training and inference workloads in production today.
- We operate with urgency, deep ownership, and a strong bias toward execution.
- Reliability, operational excellence, and rigorous systems engineering are core to our business.
- As a Sr Staff System Engineer, GPU Fleet, you will be the senior technical owner for CIC’s large‑scale GPU compute infrastructure.
- This is a hands‑on senior individual contributor role with fleet‑level responsibility and broad cross‑functional influence.
- You will define the technical direction for how GPU fleets are architected, operated, automated, and evolved across multiple generations of hardware.
- Your work will directly affect fleet reliability, operating efficiency, scalability, and customer success.
- This role does not involve people management, but it carries principal‑level scope, autonomy, and decision‑making authority across infrastructure, hardware, and operations.
- Fleet Architecture & Technical Ownership
- Own the end‑to‑end technical architecture of hyperscale GPU fleets, including hardware platform selection, firmware strategy, OS configuration, drivers, networking, and observability.
- Define and enforce technical standards and best practices for fleet reliability, availability, performance, and operability.
- Lead major fleet‑wide initiatives such as new GPU platform bring‑ups, multi‑generation hardware transitions, and architectural redesigns.
- Evaluate trade‑offs across cost, performance, reliability, and time‑to‑deploy, and make technically sound decisions under ambiguity.
- Reliability, Availability & Performance
- Set and drive fleet‑level reliability, availability, and performance objectives.
- Lead root‑cause analysis and resolution of complex, systemic failures affecting large portions of the fleet or multiple datacenters.
- Identify recurring failure patterns and drive long‑term fixes spanning hardware, software, automation, and operational processes.
- Work directly with hardware vendors and partners to resolve platform‑level issues and influence future hardware designs.
- Automation & Systems Engineering
- Design and build large‑scale automation systems for:
- GPU fleet provisioning and lifecycle management
- GPU health validation, diagnostics, and certification
- Automated remediation, recovery, and replacement workflows
- Eliminate manual operational toil through durable, well‑designed tooling that scales with fleet growth.
- Ensure all fleet systems are observable, testable, and resilient under failure conditions.
- Operational Leadership Act as a senior escalation point for critical production incidents impacting GPU availability or customer workloads.
- Participate in on‑call rotations with a strong emphasis on preventing future incidents, not just responding to them.
- Lead high‑severity post‑incident reviews and ensure learnings are translated into concrete engineering and process improvements.
- Technical Influence & Mentorship
- Provide technical mentorship and guidance to system and infrastructure engineers across the organization.
- Serve as a trusted technical partner to platform engineering, networking, datacenter operations, and leadership teams.
- Influence CIC’s long‑term infrastructure roadmap through strong technical judgment and data‑driven recommendations.

More openings at Coupang