Software Development Engineer Location Technologies, Sensing & Connectivity

Our mission is to personalize the user experience on Apple devices based on where you go when and what those places mean to you. Youre experiencing our work whenever you see a suggested location in Maps or Calendar or browse your Memories in Photos or Journal. Were working for you whenever your phone engages Do Not Disturb While Driving or remembers where you the Location Context team and we build the location intelligence backbone powering Maps Visited Places Siri location suggestions and predictive features across the OS. Were looking for engineers who love solving hard problems at the intersection of location state estimation on-device machine learning and privacy-preserving you excited by any of these challengesn Building location state estimators that fuse GPS WiFi IMU and altimeter data to understand not just where users are but what floor of a building theyre onn Designing ML models to infer the semantics of a place and forecast where the device will go next entirely on-device with strict power and memory budgetsn Developing clustering algorithms and data pipelines that process billions of location events while preserving user privacyn Optimizing system performance at massive scalewhere a 1% edge case impacts 10 million devices and a power regression of 0.1% mattersn Collaborating with Maps Siri Photos HomeKit Journal and Safety teams to power features that require deep contextual understandingnIf this sounds like you read on.

In this role youll develop the next frontier of location intelligence in partnership with teams across sensing Siri Maps and system frameworks. Youll work on problems from research through production deployment:nDesign and implement location state estimation algorithms that fuse multi-modal sensor data (GPS WiFi positioning accelerometer altimeter barometer) to build a rich understanding of user context and mobility patternsnDevelop on-device machine learning models for place inference route prediction and behavioral forecasting that operate within strict power and memory constraintsnBuild data processing pipelines that aggregate filter and cluster real-world sensor data on mobile devices balancing intelligence with resource constraintsnImplement sophisticated algorithms for background location awareness and semantic understanding then integrate them into production code running on hundreds of millions of devicesnCollect and analyze real-world datasets to train models validate performance and iterate on algorithm designnTest rigorously. Dogfood your work. Collect metrics across diverse user populations and edge cases. An issue that affects 1% of a billion devices is a big for the full system: CPU memory power consumption and radio usage. Our software needs to provide a high level of intelligence while sipping batterythis is one of the most exciting engineering challenges in mobile dedication to users privacy and security is core to how Apple does business. We want their devices to exhibit the high level of intelligence and proactivity that can only come from deep contextual understanding. We dont want their sensitive data coming back to Apple or being exposed to third parties. Other companies solve similar problems in very different ways. Our way is more work. We believe its worth it.n

Conceptualize explore and define new inferential and predictive location- and motion-based capabilities for Apples platformsnDesign and implement location state estimation algorithms sensor fusion techniques and ML models for on-device inferencenDevelop clustering and pattern recognition algorithms to identify significant locations routes and behavioral patterns from noisy sensor datanBuild and optimize data processing pipelines that operate within strict power and memory budgets on mobile hardwarenCollect curate and analyze real-world datasets of varying size and complexity to validate algorithm performancenIntegrate algorithms into production code (Objective-C Swift C) working within daemon and framework architecturesnProfile and optimize system performance: measure CPU memory footprint power consumption and latency; iterate to improvenCollaborate across teams (Maps Siri Photos Health Safety) to understand requirements and deliver capabilities that enable compelling user experiencesnWrite robust maintainable code. Test thoroughly. Address edge cases. Build systems that scale to billions of devices.

5 years experience developing commercial software preferably systems-level or embedded software running on resource-constrained devicesnStrong programming skills in C C Objective-C or Swift with solid foundation in algorithms data structures and computational complexitynWorking knowledge of statistics and probability including comfort with histograms probability distributions Bayesian inference and hypothesis testingnExperience evaluating and optimizing system performance: memory footprint CPU usage power consumption and I/O

Deep expertise in location technologies: GPS/GNSS positioning WiFi-based localization indoor positioning sensor fusion for state estimation or IMU-based dead reckoning. If youve built location estimators that fuse multiple sensor modalities we especially want to hear from with machine learning for time-series data spatial data or behavioral prediction. On-device ML experience (model size optimization quantization power-efficient inference) is a strong in signal processing Kalman filtering particle filters or other probabilistic state estimation with clustering algorithms (DBSCAN hierarchical clustering etc.) and unsupervised learning applied to spatial or temporal record of shipping production systems that operate at scale under resource constraints (mobile embedded or edge computing environments).nStrong collaboration skills and ability to work effectively across teams with diverse expertise. At Apple youll partner closely with teams in sensing connectivity privacy and application frameworks. Youll need to communicate clearly plan collaboratively and execute with performance profiling tools (Instruments dtrace etc.) and systematic optimization of CPU memory and power with large-scale data analysis for offline algorithm development model validation and performance evaluation across diverse user populations.