Principal Product Manager/Architect - Foundry Inference Platform (CoreAI)

We are seeking a Principal Product Manager/Architect to define and guide the technical architecture of Microsoft Foundry as the most reliable, scalable, and efficient AI inferencing platform in the industry. This role sits at the intersection of platform architecture, largescale GPU fleet management, and strategic customer engagement, with end-to-end accountability for the product direction that shape reliability, efficiency, and customer trust at global scale. 

This leader will partner with Engineering and Product Management leaders to drive reliability, efficiency and strategic customer engagement while remaining deeply engaged in nearterm execution. The role partners closely with engineering, product, and customer teams across CoreAI, Azure, and 1P products to ensure Foundry delivers industryleading reliability, worldclass GPU efficiency, and differentiated value for Microsoft’s most strategic AI customers. 

1. Product Reliability 

Own the product direction for Microsoft Foundry inference, with a primary mandate to make the platform the most reliable enterprise inferencing service available. This includes defining architectural standards for global serving, multi-region resiliency, automated failover, and platform-managed disaster recovery, evolving the system from customer-managed resilience to platform-managed global reliability.  

Drive architectural alignment across global routing, capacity pooling, observability, and control plane abstractions to ensure consistent availability, predictable recovery behavior, and simplified customer operations at scale. Partner with engineering, infrastructure, and security leaders to ensure reliability targets, SLAs, SLOs and recovery objectives are designed into the platform by default, not added as afterthoughts.  

2. GPU Fleet Efficiency & Capacity 

Set the product direction for GPU fleet efficiency and capacity management, guiding platform-level design decisions that maximize utilization, minimize fragmentation, and accelerate timetomonetization of new hardware and models. 

This includes shaping the architecture for global capacity pooling, intelligent scheduling, fungibility across workloads, automated demand forecasting, and softwaredefined allocation, ensuring Foundry can scale demand while operating within realworld supply constraints. The role will work closely with efficiency and infra teams to translate deep optimization learnings into durable platform primitives, enabling sustained efficiency gains rather than oneoff wins.  

The Product Manager/Architect is expected to influence architectural investments across inference utilization, model serving, and hardware/system performance, ensuring that efficiency improvements are systemic, measurable, and repeatable across generations of models and GPUs.  

3. Strategic Customer & Innovation Engagement 

Act as a senior technical advisor and architect for Foundry’s most innovative and strategic customers, particularly those pushing the boundaries of scale, reliability, or model complexity. Engage directly with customers on deep technical challenges, including largescale model migrations, reliabilitysensitive production deployments, and advanced serving architectures. 

Support competitive and strategic initiatives by articulating Foundry’s architectural advantages, turning bespoke requests into scalable features, and ensuring customer feedback meaningfully influences platform roadmap and architectural priorities. This role will frequently operate at CTO/Chief Architect level with customers, translating complex platform internals into clear, credible architectural guidance. 

4. Cross-Company Technical Leadership 

Serve as a unifying architectural voice across product management, engineering, infrastructure, and partner teams. Drive alignment on longterm technical direction, resolve architectural tradeoffs, and provide clear guidance on when to optimize for reliability, efficiency, performance, or speed. 

The Product Manager/Architect will regularly engage with senior Microsoft leadership across 1P teams, producing architectural briefs, decision frameworks, and recommendations that connect technical design choices to business outcomes, including cost efficiency, customer trust, and platform differentiation.  

What are the success metrics for this role? 

• Platform Reliability & Trust:
  • Reduction in customer-visibility reliability incidents
  • Adoption of platform-managed resilience primitives
  • Reduce effective RTO/MTTR 

• Deliver platform scale without customer perceived capacity constraints
  • Improve revenue/GPU & Time to Revenue
  • Reduce fragmentation across SKU, region & workload
  • Self-Service success rate improvement for scale-up/scale-down
• Strategic Customer