Principal Architect — AI Cloud & GPU Infrastructure

CloudData.Center; USA; Remote; USD 180.000 - 270.000; Vollzeit

Zusammenfassung

CloudData.Center is seeking a senior architect to set the technical direction for AI compute environments, spanning GPU clusters, private AI clouds, and MLOps platforms. The role blends architectural leadership with hands-on design and deployment for client infrastructure.

You will lead GPU cluster designs (H100, B200, GB200), plan high‑performance networking (InfiniBand, 400G/800G Ethernet), and shape architectures for inference serving and AI workloads.

Qualifikationen

• Deep expertise in AI compute and GPU infrastructure architecture.
• Experience designing GPU clusters and high-performance networking for AI workloads.
• Familiarity with Kubernetes-based platforms and inference serving.

Aufgaben

• Lead designs for GPU compute clusters using systems in the H100, B200, and GB200 classes.
• Plan high-performance AI networking, including InfiniBand and 400G/800G Ethernet fabrics.
• Shape private AI cloud and MLOps platform architectures.
• Design infrastructure for inference serving and related AI workloads.
• Guide deployment choices across compute, networking, and platform layers.

Kenntnisse

AI compute architecture

GPU cluster design

High-performance networking

Kubernetes platforms

Inference serving

Tools

InfiniBand networking

400G/800G Ethernet

Jobbeschreibung

Role overview

Set technical direction for AI compute environments, from GPU clusters and private AI clouds to the platforms used to train and serve models. The role combines architecture leadership with hands-on design and deployment for client infrastructure.

Responsibilities

• Lead designs for GPU compute clusters using systems in the H100, B200, and GB200 classes.
• Plan high-performance AI networking, including InfiniBand and 400G/800G Ethernet fabrics.
• Shape private AI cloud and MLOps platform architectures.
• Design infrastructure for inference serving and related AI workloads.
• Guide deployment choices across compute, networking, and platform layers.

Requirements

• Deep expertise in AI compute and GPU infrastructure architecture.
• Experience designing GPU clusters and high-performance networking for AI workloads.
• Familiarity with Kubernetes-based platforms and inference serving.

Benefits and work setup

• Full-time, remote role; the listed locations are the United States or India.