Machine Learning Engineer - Model Inference

About the Role

Join Apple Maps to help build the best map in the world! In this role on our ML Platform Team, you will leverage advanced deep learning and large language models to improve the search quality and overall customer experiences across our various Maps platforms. This role offers amazing opportunities to partner closely with research and product teams while taking ownership of projects and delivering measurable results at a global scale!

Responsibilities

• Design, build, and operate the services used to deploy and serve machine learning models at scale
• Oversee the infrastructure that powers model inference, from developing high-performance serving systems to implementing optimization techniques that reduce latency, increase throughput, and improve hardware utilization
• Collaborate closely with machine learning researchers, infrastructure engineers, and product teams to transform new models into reliable, production-ready experiences
• Communicate technical ideas clearly and present design decisions and performance findings to both technical and cross-functional audiences
• Participate in collaborative discussions, design reviews, and project planning meetings
• Learn quickly, take ownership of meaningful projects, and contribute ideas that improve both the performance of systems and the experiences of users

Minimum Qualifications

• Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical field plus at least 2 years of post-graduate work experience
• Strong programming skills in Python and at least one systems-oriented language such as C++, Rust, or Go
• Solid understanding of data structures, algorithms, operating systems, and computer architecture
• Familiarity with machine learning fundamentals and modern deep learning frameworks such as PyTorch, TensorFlow, or JAX
• Experience building, debugging, or evaluating software systems through coursework, internships, research, open-source contributions, or personal projects
• Ability to analyze technical problems, communicate clearly, and work effectively with engineers across multiple disciplines

Preferred Qualifications

• Experience with model serving technologies such as Triton, TensorRT, ONNX Runtime, vLLM, TensorFlow Serving, or TorchServe
• Familiarity with inference optimization techniques, including quantization, pruning, knowledge distillation, speculative decoding, kernel fusion, or continuous batching
• Understanding of GPUs, accelerators, distributed systems, networking, or high-performance computing
• Familiarity with containers, Kubernetes, cloud infrastructure, and production observability tools
• Experience benchmarking large language models, vision models, or other compute-intensive machine learning workloads
• Curiosity about how software, models, and hardware interact to determine real-world performance