GPU Infrastructure Engineer for Scalable AI Inference

AI Startups UK; Greater London; Hybrid; GBP 120,000 - 180,000; Full time

Job summary

OpenAI is seeking a software engineer to build and operate large-scale production systems that power ChatGPT's GPU fleet. You will develop tooling for fleet health, capacity planning, automation, and incident response, working with infrastructure, research, and product teams.

Candidates should have 5+ years in software engineering, strong skills in Go, Python, C++, or Rust, and experience with distributed systems and GPU infrastructure.

Qualifications

• Five or more years of software engineering experience building production infrastructure.
• Strong programming skills in Go, Python, C++, or Rust.
• Experience designing or operating highly available distributed systems.
• Experience with GPU infrastructure or large-scale compute platforms.
• Strong debugging and operational problem-solving skills.
• Excellent cross-team communication.

Responsibilities

• Build software and internal tools to manage large-scale GPU infrastructure powering ChatGPT inference.
• Develop systems for capacity planning, fleet health monitoring, and resource utilization.
• Automate operational workflows, including incident detection, diagnosis, and response.
• Identify bottlenecks affecting fleet reliability, scalability, and performance.
• Partner with infrastructure, research, and product engineering teams to improve the compute platform.

Skills

Go

Python

C++

Rust

Distributed systems

Production infrastructure

Debugging

Communication

Tools

GPU infrastructure

Cluster orchestration

Job description

OpenAI is seeking a software engineer to build and operate large-scale production systems that power ChatGPT's GPU fleet. You will develop tooling for fleet health, capacity planning, automation, and incident response, working with infrastructure, research, and product teams.

Candidates should have 5+ years in software engineering, strong skills in Go, Python, C++, or Rust, and experience with distributed systems and GPU infrastructure.