Apple is looking to build the next generation of tools and platforms that power high-impact decisions across Siri, specifically focusing on improving Siri's User Experience by automatically identifying regressions in Siri performance and alerting engineering teams with actionable insights for debugging performance issues.
Requirements
- Strong understanding of core ML concepts, with particular depth in unsupervised learning methods (clustering, dimensionality reduction, density estimation), and a solid foundation in feature engineering, model evaluation, regularization, and optimization.
- Advanced coding skills in Python (5+ years) with pandas, scikit-learn, and at least one deep learning framework (PyTorch or TensorFlow).
- Hands-on experience data preprocessing, building and training ML models using distributed processing frameworks such as PySpark, Spark, or Flink.
- Experience applying large language models (LLMs) for downstream tasks (classification, labeling, enrichment), with the ability to perform fine-tuning or parameter-efficient adaptation (e.g., LoRA).
- Must be capable of deploying and optimizing models in on-premise, server, or on-device environments, rather than relying solely on hosted third-party APIs
- Proven expertise with anomaly detection and time series modeling (e.g., Isolation Forest, autoencoders, ARIMA, LSTM) and experience building production frameworks supporting multiple engineering and product teams.
- Experience with LLM workflows (domain adaptation, RAG) and deploying optimized ML/LLM models on mobile or server environments (e.g., Core ML, TensorFlow Lite, ONNX Runtime) for performance, cost, and privacy.
Responsibilities
- Design and implement scalable, reliable systems to transform raw data into actionable insights for leadership.
- Building a trustworthy and explainable anomaly detection system to automatically identify regressions in Siri performance and alert engineering teams with actionable insights for debugging performance issues.
- Drive the technical vision for Siri’s automated anomaly detection platform for detecting performance and reliability regressions.
- Defining, developing and delivering key features for high quality alerting to enable teams to troubleshoot regressions rapidly.
- Gathering feature requirements, defines technical roadmaps and executes efficiently.
- Owning the technical roadmap, onboarding and mentoring team members, and leading the team to deliver high-impact outcomes.
- Executing in a rapidly changing environment with ambiguous requirements to drive impact incrementally.
Other
- Operate at the intersection of engineering excellence and business impact.
- Demonstrates strong communication and technical leadership skills and the ability to engage with colleagues and leadership to find common ground on solving hard problems.
- Demonstrated ability to set technical vision, lead complex projects, and drive impact in cross-functional environments, with strong communication and problem-solving skills.
- Strong problem solving skills and are self-directed with a proven ability to execute.
- Continually desire learning and demonstrate attention to details and find opportunities to innovate and share knowledge with others.
