GPU Infrastructure Architect - AI Deployments

Overview

Tata Consultancy Services, a global leader in IT services and consulting with over 50 years of experience, is looking for a proficient GPU Infrastructure Architect to design and implement large-scale GPU architectures tailored to enterprise, regulated, and sovereign AI projects. This role involves collaboration with top technology partners and optimizing high-performance GPU clusters.

Key Responsibilities

• Architect comprehensive GPU infrastructure solutions for advanced AI workloads across enterprise and regulated sectors.
• Develop sizing calculators and layered reference architectures, specifically for H100, H200, and B200 GPU platforms.
• Lead efforts in cluster validation, performance benchmarking, system readiness checks, and tuning for peak efficiency.
• Provide expertise on secure deployment scenarios such as air-gapped environments, virtual private cloud isolation, registry mirroring, and on-premise NIM hosting.
• Collaborate extensively with NVIDIA, hyperscalers, and OEM partners to finalize architectural choices and address technical challenges.
• Create technical documentation including whitepapers, reference architecture presentations, total cost of ownership (TCO) models, and enablement resources.

Required Qualifications

• Graduation degree in any discipline.

Skills and Tools

• Proficiency in designing GPU cluster architectures focusing on NVIDIA hardware like H100, H200, B200 GPUs.
• Experience with AI deployment patterns in enterprise and secure environments.
• Capability in benchmarking and tuning high-performance clusters.
• Familiarity with network isolation techniques such as VPC and air-gapped systems.
• Strong collaboration skills with hyperscalers and original equipment manufacturers (OEMs).
• Technical writing skills for developing whitepapers and architectural documentation.