Ai Infrastructure Principal Architect
Pubblicato il 19-09-2026 - Accenture in Bardi
YOU ARE As a Principal AI Infrastructure Architect, you are the firm's most senior technical authority on compute infrastructure for large-scale AI and machine learning systems, bringing extensive experience as a lead and senior architect along with command over a broad landscape of technological options and the latest innovations that can be introduced into a solution. You have a proven track record of successfully designing and deploying large-scale infrastructure on which significant AI/ML solutions operate in production and demonstrably deliver business value — not just systems that perform, but systems that move the needle for the organizations they serve. You weigh and rationalize multiple viable architectures across compute, networking, storage, orchestration, and model serving, making authoritative decisions tailored to each client's situation, standards, and strategic objectives, and you set the technical direction that senior and lead architects build upon. As a recognized expert across at least one hyperscaler cloud — and conversant across several — you bring deep,
current knowledge of AI/ML services, accelerators, interconnects, and cost levers, and you continuously scan the horizon to identify promising emerging technologies and judge where and when they belong in a real solution. You are well known to our infrastructure partners and partnering organizations, maintaining strong relationships that give the firm early access, influence, and insight, and you represent the practice's technical credibility in those forums. Beyond architecture, you shape strategy and roadmaps, establish standards and reference architectures, mentor and elevate the architect community, and are ultimately accountable for ensuring the firm delivers AI/ML infrastructure that meets business SLAs, controls cost, scales to frontier workloads, and creates lasting business value. THE WORK Set the overarching technical vision and strategy for compute infrastructure supporting large-sca