Applied Machine Learning Engineer

Quick Facts

• Role: Applied Machine Learning Engineer (Location Technologies)

• Focus: Production ML systems for location context and device behavior inference

Description

You will lead feature direction and technical strategy for location intelligence, partnering with engineers and researchers across sensing, Siri, and apps. You’ll translate ambiguous product goals into a multi-year technical roadmap and make architectural trade-offs. You’ll design, build, evaluate, and iterate on production ML systems that infer device patterns from GPS, Wi‑Fi, accelerometers, and higher semantic signals, while tracking metrics and improving quality.

Responsibilities

• Set feature development direction and technical strategy for location technologies

• Collaborate across engineering, product, and research leadership to define multi-year roadmap

• Design, build, evaluate, and refine production ML systems for contextual inference

• Use estimation techniques and machine learning to infer behavior patterns from sensor and context data

• Test, measure, and iterate based on quality metrics; use the systems yourself

• Identify new opportunities for location intelligence across products

• Mentor engineers and raise the technical bar

Requirements

• Experience building and shipping production software, with technical leadership of multi-engineer projects

• Experience shipping production software for mobile and/or other resource-constrained devices

• Ability to create, analyze, and modify software functionality, ideally in C++/Obj-C/Swift codebases

• Skilled working with large, noisy datasets

• Experience with NumPy, pandas, scikit-learn, and PyTorch or TensorFlow

• Hands-on experience with applied probability, statistics, and empirical and/or ML algorithms (estimation, signal processing, or supervised ML)

• Bachelor’s or graduate degree in Computer Science, Computer Engineering, Mathematics, or related field

Benefits

• Opportunity to build impactful, customer-facing location intelligence within tight CPU, memory, and schedule constraints