Hp AI Runtime / ML Model Systems Engineer to define architecture and performance strategy for AI workloads across HP PCs. Focus on runtime abstraction, model placement, and execution strategies across CPUs, GPUs, and NPUs.
Responsibilities
We are looking for an AI Runtime / ML Model Systems Engineer to define the architecture and performance strategy for AI workloads across HP PCs. You will determine how software should use compute engines, memory, hardware capabilities, and different model types to deliver reliable and responsive AI experiences. Working with platform architects and engineering teams, you will shape runtime abstraction frameworks, model placement policies, and execution strategies across CPUs, GPUs, and NPUs. You will translate product and experience requirements into architectural decisions that balance inference performance, resource availability, power consumption, and thermal constraints. This role emphasizes system design, technical planning, and architectural guidance. Success requires deep knowledge of AI runtime behavior and the ability to define solutions that engineering teams can implement and validate across different hardware configurations.
What You Might Do
Define runtime architectures and performance strategies for AI workloads across heterogeneous compute resources.
Assess how model types, hardware capabilities, memory requirements, and workload behavior should influence system design.
Design runtime abstraction layers and interfaces that connect applications and models to inference engines and hardware backends.
Define model placement, scheduling, and routing policies across CPU, GPU, and NPU resources, including fallback behavior and resource availability considerations.
Evaluate inference technologies such as ONNX Runtime, TensorRT, and OpenVINO to guide framework selection and integration decisions.
Establish approaches for memory allocation, model lifecycle management, and resource coordination when multiple AI workloads execute concurrently.
Translate product requirements into performance targets, system requirements, and architectural guidance for development teams.
Define benchmarks and validation criteria; use prototype results and system measurements to assess architectural choices.
Lead design reviews and collaborate with platform, security, and hardware and software partners to resolve dependencies and guide engineering roadmaps.
Essential Qualifications
Qualification
Knowledge of agentic applications
Preferred
Experience architecting AI runtime or system software for Windows PCs.
Familiarity with hardware acceleration backends and software ecosystems from Intel, AMD, NVIDIA, or Qualcomm.
Knowledge of different model architectures, quantization approaches, and their implications for compatibility, memory footprint, and execution.
Experience planning for resource contention, power and thermal constraints, and concurrent AI workloads.
Familiarity with C++ and Python for evaluating technical feasibility, reviewing implementations, or developing architectural proofs of concept.
Knowledge of agentic applications, multiple-model execution, and local and cloud inference strategies.
