On-Device AI Runtime Engineer

A globally leading technology company is looking for an On-Device AI Runtime Engineer to join its cutting-edge AI team. In this position, you will be a key contributor in building high-performance machine learning inference systems, developing optimized AI drivers for edge devices, creating scalable model lifecycle management solutions, and delivering efficient on-device runtime drivers for AI inference across a wide range of hardware platforms. This role will focus on optimizing and deploying machine learning models on edge and mobile devices. Please note that this is not a Data Science or Data Analyst role. Job Responsibilities: • Design and implement robust Core ML model optimization pipelines for deploying large-scale ML models on resource-constrained devices. • Support product engineering teams by consulting on AI model performance, iterating on inference solutions to solve real-world mobile/edge AI problems, and developing/delivering custom on-device AI frameworks. • Interface with hardware and platform teams to ensure optimal utilization of neural processing units (NPUs), GPUs, and specialized AI accelerators across the device ecosystem. Minimum Qualifications: • Strong proficiency in Swift/Objective-C and Metal Performance Shaders. • Familiar with various ML model formats such as Core ML, ONNX, TensorFlow Lite, and PyTorch Mobile. • Strong critical thinking, performance optimization, and low-level system design skills. • Experience with model quantization, pruning, and hardware-aware neural architecture optimization. • Experience with real-time inference pipelines and latency-critical AI applications. • Understanding of mobile device thermal management, power consumption patterns, and compute resource allocation for AI workloads. Type: Contract Duration: 3 months (with a possibility for extension) Work Location: San Diego or Sunnyvale, CA (On-site) Pay rate: $78.00 - $93.00 (DOE)