Apple is looking to solve the business and technical problem of enabling the Research to Production lifecycle of innovative machine learning models that power user experiences on Apple's hardware and software platforms, specifically focusing on the Authoring and Conversion toolchain for onboarding ML models into Apple's ML deployment platform.
Requirements
- Experience with MLIR / LLVM compiler technologies
- Experience with on-device ML frameworks (Core ML, Win ML, ONNX, TF Lite or ExecuTorch)
- Knowledge of ML development and workflows, including at least one authoring framework experience (e.g., PyTorch).
- Solid programming skills in Python, C++ or Swift.
- Deep technical understanding of ML research production workflows, ML compilations, runtimes, and on-device optimization techniques (such as quantization, distillation, etc.).
- Familiarity with Apple framework and API design patterns
Responsibilities
- Lead the team building the authoring/conversion technologies to quickly onboard new ML models to our ML platform, including contributions to internal and external ML authoring frameworks, like MLX and PyTorch.
- Understand different ML operations, architectures, and graph representations in different authoring frameworks and stay on top of the latest innovations in this space.
- Partnering with multi-functional partners to push the state-of-the-art of on-device ML functionality and performance.
- Planning and implementing multi-year roadmaps for both internal architectures and public facing functionality.
- Engaging with and supporting 1st and 3rd party developers ML powered use cases.
- Develop the foundational authoring and conversion APIs that enable the authoring and import of ML models into Apple’s ML deployment platform.
- Owns the Python authoring toolchain and APIs, the MLIR front-end compiler and intermediate representations, and the integration with ML research frameworks.
Other
- Two or more years of strong and validated management experience.
- Attracting, hiring, and guiding the career of ML and software engineers.
- Excellent social skills.
- Track record of creating clean software architectures, intuitive designs, and high-performance extensible software.