Summary
The Siri Conversation Intelligence (SCI) team is crafting groundbreaking technology for large scale systems, spoken language, machine learning, and LLMs to enable better conversational experience with Siri. The features we build are redefining how hundreds of millions of people use their computers and mobile devices to fulfill their requirements and find what they are looking for. As part of the SCI team, you will work with a hard-working team who crafted the intelligent assistant that helps users get things done - just by asking!

You'll drive research in deep learning and develop the most sophisticated natural language and conversation understanding systems. You should be passionate about working with software and platform engineers to integrate the models on-device and improve the natural language understanding (NLU) ability of Siri. Excellent interpersonal skills will be required to coordinate work across multiple teams to ship the state-of-the-art technology to production and impact all Siri users.

Key Qualifications
Proven understanding of machine learning technology, deep learning (including LLMs) and NLU

5-10+ years proven programming skills using standard ML tools such as C/C++, Python, PyTorch, Tensorflow, Hugging Face, etc.

Hands-on experience working (training, fine-tuning, optimizing, deploying) with large models (e.g. LLMs)

Hands-on experience applying common machine learning optimization techniques, like quantization and distillation, to reduce the resource consumption and/or eliminate latency

Excellent problem-solving (e.g. via building forward-looking prototype systems), critical thinking, strong communication, and collaboration skills

Publication at top ML/DL/NLP conferences such as NeurIPS, ACL, EMNLP, etc. is a plus!

Description
As a member of this team, you'll develop novel ML/DL models including LLMs for NLU and conversational understanding, work with platform and software engineers to convert and compile the models to run on device and integrate them with runtime systems, optimize latency and performance of the models, and diagnose model errors and perform accuracy hill-climbing for shipped products, etc.

Education & Experience
Master's, or PhD or equivalent experience in Computer Science, Machine Learning or a related field

---