About HP

At HP, you’ll have a chance to create tools, technology, and solutions that reshape the way the world works in the future. If you’re looking to join a company that allows you to connect with a network of professionals eager to support you in doing your best work, we want to talk to you. Our legendary culture guides every employee toward success—fostering collaboration and driving innovation.

Operating in over 170 countries, HP is always creating new services, products, and capabilities giving you more opportunities to advance your career. Here, innovation is the key to professional development and career mobility.

When you join HP, you’re joining a company that believes every voice matters and that we all deserve a seat at the table. From the boardroom to factory floor, we create a culture where everyone is respected and where people can be themselves. You will be part of a global laboratory where different perspectives and experiences will help you solve problems in new ways. This is where you can build a long and wide-ranging career.

About The Role

We are looking for a Principal AI Runtime & Performance Lead to define the architecture and performance strategy for AI workloads across HP PCs. You will establish how software should use compute engines, memory, hardware capabilities, and different model types to deliver responsive, efficient, and reliable AI experiences. As a senior technical leader, you will translate product goals into system-level performance requirements and architectural direction. You will guide decisions about model execution, resource allocation, and runtime behavior across CPUs, GPUs, and NPUs, considering the constraints of different device configurations. The role emphasizes architectural planning and technical leadership. You will set the direction for implementation and optimization teams, establish how performance should be measured, and use evidence from prototypes, benchmarks, and telemetry to refine the architecture.

What You Might Do

Own the technical direction for AI runtime architecture and performance across supported PC platforms and hardware configurations.

Define model execution strategies across CPU, GPU, and NPU resources, balancing responsiveness, throughput, memory use, power consumption, and thermal constraints.

Determine how different model types and workload patterns should influence runtime design, hardware utilization, and resource allocation.

Establish architectural approaches for concurrent agents and models, including resource coordination and behavior under system pressure.

Translate customer experience goals into performance budgets, system requirements, and design principles that engineering teams can implement.

Define benchmark methodology and representative workload scenarios to evaluate architectural choices and measure progress against performance goals.

Guide cross-functional teams through design reviews, resolve technical trade-offs, and prioritize improvements based on system-level impact.

Partner with hardware, firmware, operating system, platform, and AI engineering teams to align capabilities and dependencies across the stack.

Shape engineering roadmaps, evaluate emerging technologies through proofs of concept, and mentor engineers on runtime architecture and performance design.

Essential Qualifications

10+ years of experience in software engineering, systems architecture, runtime systems, or a related field, including technical leadership of complex engineering initiatives.

Demonstrated experience defining architecture and performance strategy for AI runtimes, inference platforms, or hardware-aware software systems.

Strong understanding of heterogeneous computing, including CPU, GPU, and NPU execution and the implications for model placement and resource utilization.

Deep knowledge of memory behavior, concurrency, scheduling, and the interaction among applications, operating systems, runtimes, and hardware.

Strong Linux and Windows expertise, with the ability to assess how operating system capabilities and constraints affect runtime design.