On-Device ML Infra Engineer — API & Model Optimization

Socket.dev; Cupertino (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Apple is seeking an ML Infrastructure Engineer focused on ML user experience APIs and integration for on-device AI. You will develop new model authoring and conversion APIs that serve as entry points into Apple’s ML infrastructure and drive onboarding of popular models into Apple’s stack with strong, competitive performance on Apple devices.

You will collaborate across teams to integrate these APIs into internal and external model repositories, optimize pipelines from authoring to runtime, and

Qualifications

• Bachelor's degree in Computer Science, Engineering, or related field.
• Proficiency in Python; familiarity with C++ required.
• Experience with ML frameworks such as PyTorch, MLX, and JAX.
• Strong understanding of ML fundamentals including Transformer architectures.
• Hands-on experience with ML inference optimizations (quantization, pruning, KV caching).
• Strong communication skills across multi-functional teams.

Responsibilities

• Develop and expose APIs that enable ML engineers to author and convert models for Apple platforms.
• Integrate ML tools into internal and external repositories (e.g., Hugging Face) to showcase end-to-end workflows.
• Implement optimizations and transformations from model authoring to runtime execution across Apple hardware.
• Collaborate on building the end-to-end ML inference stack, including profiling, debugging, and analysis.

Skills

Python

C++

PyTorch

MLX

JAX

Transformers

Quantization

Pruning

KV caching

Communication

Education

Bachelor's degree in Computer Science or related

Tools

Hugging Face

MLIR/LLVM

Job description

Apple is seeking an ML Infrastructure Engineer focused on ML user experience APIs and integration for on-device AI. You will develop new model authoring and conversion APIs that serve as entry points into Apple’s ML infrastructure and drive onboarding of popular models into Apple’s stack with strong, competitive performance on Apple devices.

You will collaborate across teams to integrate these APIs into internal and external model repositories, optimize pipelines from authoring to runtime, and