Member of Technical Staff - GPU Infrastructure Engineer

About the role

Liquid AI is seeking a hands‑on engineer to ensure the reliability and efficiency of its GPU clusters that power foundation model training and research. The role focuses on reducing downtime, improving resource utilization, and building tools that let researchers concentrate on model development.

Key responsibilities

• Own the operation and reliability of GPU clusters used for large‑scale AI training.
• Debug and resolve issues across compute, storage, networking, schedulers and distributed workloads.
• Increase CPU, GPU and storage utilization through automation and better tooling.
• Onboard and migrate workloads across different GPU providers and hardware platforms.
• Develop monitoring, validation and platform abstractions that reduce operational overhead for researchers.
• Contribute to the long‑term architecture of Liquid AI’s training infrastructure and GPU platform.

Required profile

• Strong software engineering background with experience building production‑grade infrastructure tools and automation.
• Deep understanding of distributed systems, Linux, networking and storage.
• Hands‑on experience operating shared compute clusters or distributed training platforms.
• Proven track record of supporting production users and turning recurring failures into durable solutions.
• Ability to collaborate effectively with senior research and infrastructure engineers.

Required skills

• Linux
• Networking
• Storage
• Distributed systems
• SLURM
• Kubernetes
• Ray
• Hadoop
• GPU
• HPC