Principal Machine Learning Researcher, On-Device Optimization

About This Role

Advance on-device AI optimization techniques, bridging applied research and product development at HP IQ, an AI innovation lab. Develop and implement model compression techniques to deploy state-of-the-art transformer and vision models on HP laptops and edge devices.

Responsibilities

• Research and implement model compression techniques including quantization, low-rank factorization, distillation, and pruning
• Develop methods to deploy SOTA transformer and vision models on-device under hardware constraints
• Lead investigations into hardware-aware training strategies to optimize latency, throughput, and memory usage
• Collaborate with software engineers and system architects to integrate models into AI companion apps
• Evaluate and benchmark different frameworks and quantization strategies (e.g., AWQ, GPTQ, SmoothQuant)

Requirements

• PhD in Computer Science, Electrical Engineering, or related field with focus on efficient ML, systems ML, or compiler design for ML
• 2+ years of industry or applied research experience
• Strong background in model optimization for edge computing or mobile/embedded deployment
• Familiarity with PyTorch, ONNX, TensorRT, OpenVINO, QNN, or Llama.cpp
• Understanding of tradeoffs in asymmetric/symmetric quantization, calibration methods, and inference tuning

Qualifications

• PhD in Computer Science, Electrical Engineering, or related field with focus on efficient ML, systems ML, or compiler design for ML
• 2+ years of industry or applied research experience

Nice to Have

• Experience publishing at top ML/Systems conferences (e.g., NeurIPS, ICML, MLSys)
• Familiarity with embedded ML for consumer devices
• GPU and system-level profiling tools (e.g., CUDA, nvprof, perf)
• Contributions to open-source ML optimization frameworks

Skills

* Required skills

Benefits