Machine Learning Systems Engineer, Siri Agent Modeling

What you'll do

• Optimize model training and inference performance across the full machine learning stack.
• Improve model execution speed and efficiency specifically on custom Apple Silicon hardware.
• Write production-level code to train and deploy models for Siri's AI features.
• Implement LLM optimization techniques such as quantization, KV Cache, and Speculative Decoding.
• Manage the end-to-end machine learning lifecycle including training, evaluation, and deployment.
• Utilize parallel training methodologies like FSDP and DDP to scale model development.
• Develop and deploy large language models using frameworks like PyTorch or HuggingFace.

What we're looking for

• Bachelor's degree in Computer Science, Engineering, or a related field, or equivalent experience.
• Experience in the model lifecycle including training, evaluation, and deployment.
• Strong understanding of Machine Learning architectures like Transformers and CNNs and the ML training loop.
• Proficiency in Python and machine learning frameworks such as PyTorch.
• Experience with agentic AI-assisted coding.
• Expertise in LLM optimization techniques such as quantization, KV Cache, and Speculative Decoding.
• Familiarity with training methodologies including FSDP, DDP, and other parallelism techniques.
• Proficiency in a compiled programming language like Swift, C/C++, or Java.