On-device ML Performance Engineer, Graphics, Games and Machine Learning
Summary
Analyze and optimize ML inference performance (latency, memory, power) on Apple Silicon, working across the SW stack, low-level drivers, and HW debug involving CPU, GPU, and Neural Engine using Python/C++ and frameworks like PyTorch and MLX.
The On-Device Machine Learning team at Apple is responsible for enabling the Research to Production lifecycle of cutting edge machine learning models that power magical user experiences on Apple’s hardware and software platforms. Apple is the best place to do on-device machine learning, and this team sits at the heart of that discipline, interfacing with research, SW engineering, HW engineering, and products.
The On-device ML Performance team has the responsibility to analyze latency, memory, power and numerical correctness of the latest machine learning models running on Apple SoC’s, and to make Apple’s ML software stack take full advantage of the capabilities in Apple’s ML accelerators. The work from this cross functional team enables model developers’ decisions to optimize performance via advanced techniques such as quantization, sparsity, performance and accuracy tradeoffs. The work of this team impacts all new Apple HW and ML Inference on them.
Our group is looking for an On-device ML Performance Engineer, with technical expertise in computer architecture, performance, memory, power, ML model architectures and on-device ML inference. The role entails deep analysis of ML inference from the SW stack and low level drivers to HW debug involving CPU, GPU, Apple Neural Engine, system memory and power.
As an engineer in this role, you will be primarily focused on analyzing and optimizing the performance of the latest ML models on the latest iPhones and Mac’s. You will work with models created by the most popular ML frameworks (PyTorch, MLX, etc) and will analyze the inference of those models on device to ensure the stack achieves full machine performance on Apple Silicon. The role also includes scripting, coding, and generation of utilities and debug tools to extract, analyze, and report performance and power related metrics for Apple HW. The ideal candidate will have a passion for ML model architectures and ML inference, deep knowledge of GPU and CPU, computer architecture and memory, compilers and HW drivers.
Minimum Qualifications
- Experience with ML inference, quantization, performance and accuracy
- Familiarity and experience with the most popular ML architectures (e.g. LLM’s, Diffusion models, CNN’s)
- A passion to explore and learn about the latest advances in ML model design and architecture, particularly as related to model implementation on HW and on-device inference
- Familiarity with Operating Systems, embedded systems, and CPU/GPU HW architectures
- Highly proficient in Python/C++ and shell scripting
- Familiarity with Linux or macOS
- Exceptional clarity in verbal and written communication, including the ability to present and lead discussions in larger groups
Preferred Qualifications
- Masters or PhDs in Computer Science or relevant disciplines.
- Experience with Apple’s CoreML, MPS Graph, Metal Performance Shader’s or MLX frameworks
- Experience with any on-device ML stack, such as TFLite, ONNX, ExecuTorch, etc.
- Experience with any ML authoring framework (PyTorch, TensorFlow, JAX, etc.).
- Experience with Apple’s App development framework such as Xcode, Swift, Objective-C
- Experience with any compiler stack (MLIR/LLVM/TVM etc.)