Staff ML Engineer - Inference & Distributed Systems

Apple, Inc.; Cupertino (CA); On-site; USD 216,000 - 325,000; Full time

Benefits offered by this job

Medical and dental coverage

Retirement benefits

Discounted products and services

Education expense reimbursement

Bonuses or commissions

Relocation

Job summary

Apple is seeking a Staff Machine Learning Engineer in Cupertino to own and build end-to-end ML inference systems, scalable infrastructure, and data processing flows. You will lead cross‑functional work with modeling, infrastructure, and product teams to deploy production-grade ML capabilities across Apple devices and cloud services.

You’ll drive architectural decisions, mentor engineers, and extend open‑source and Apple-developed tech to support large-scale model deployment, fine‑tuning, and

Qualifications

• Experience shipping production ML systems or ML-powered applications, with ownership of significant architecture and implementation decisions.
• Deep expertise in at least one relevant area, such as model inference, distributed computing, ML infrastructure, or data systems for ML.
• Strong programming and software engineering skills, including system design, API design, testing, and debugging complex software.
• Practical experience integrating ML models into software systems and understanding their execution and deployment requirements.
• A track record of leading technical initiatives across teams, resolving architectural tradeoffs, and delivering working systems.
• Experience investigating performance and reliability problems and driving improvements based on measured results.
• Ability to communicate technical decisions clearly, build alignment across disciplines.
• Bachelor’s degree in Computer Science, Computer Engineering, Information Systems, or Electronic Engineering and 12+ years of relevant work experience.

Responsibilities

• Build inference systems for product workloads. Develop services and APIs for model execution, including request routing, batching, concurrency, and scaling. Make tradeoffs among latency, throughput, memory use, and compute cost.
• Develop shared ML infrastructure. Adopt and extend open‑source and Apple‑developed technologies where appropriate, to build reusable components across models and applications.
• Scale fine‑tuning and training workflows. Join with modeling teams to improve distributed job execution, data loading, checkpointing, failure recovery, and accelerator utilization; iterate on foundation models and custom-built models.
• Build data-processing workflows for ML. Use distributed processing to join, filter, deduplicate, and transform large datasets into reproducible inputs for training and batch inference; address bottlenecks in data movement.
• Connect ML systems with product software. Design interfaces and implement integrations across inference services, backend applications, and on-device components, working with iOS and device-to-cloud teams.
• Set technical direction and deliver it. Turn ambiguous requirements into architecture and implementation plans, align teams on interfaces and ownership, and lead projects from prototyping through production.
• Raise engineering quality through hands-on leadership. Build critical components, diagnose issues across data, model execution, and application boundaries, and mentor engineers through design reviews and implementation.

Skills

Production ML systems

Distributed computing

API design

Software engineering

Leadership across teams

Performance & reliability

Communication of decisions

Education

Bachelor’s degree in CS/Engineering

12+ years experience

Tools

Ray / Ray Serve

ONNX Runtime

Core ML

Spark

Python

LangGraph / MCP

Job description

Apple is seeking a Staff Machine Learning Engineer in Cupertino to own and build end-to-end ML inference systems, scalable infrastructure, and data processing flows. You will lead cross‑functional work with modeling, infrastructure, and product teams to deploy production-grade ML capabilities across Apple devices and cloud services.

You’ll drive architectural decisions, mentor engineers, and extend open‑source and Apple-developed tech to support large-scale model deployment, fine‑tuning, and