Cambridge Internship in ML Model Optimization
8 Months ago • Upto 1 Years
Devops
Job Description
This internship within Microsoft's Azure Hardware Systems & Infrastructure (AHSI) organization, specifically the Strategic Planning and Architecture (SPARC) team, focuses on model compression and optimization for Large Language Models (LLMs). The intern will research and develop quantization flows for LLM inference and training, design, implement, and evaluate the performance of quantized SOTA LLMs, and present findings in technical documents and presentations. The role involves hands-on experience with quantization of LLMs, model compression, low-precision data types, and potentially PyTorch and Python. The internship is located in Cambridge, UK and is open to Masters/PhD students in Computer Science/Machine Learning or related fields.
Good To Have:
- PyTorch
- Python
- SW Tool development experience
- Excellent communication skills
Must Have:
- Masters/PhD in CS/ML
- Experience in LLM quantization and model compression
- Knowledge of low-precision data types
- Research and development of quantization flow
- Performance evaluation of quantized LLMs
- Technical document and presentation writing
Perks:
- Industry leading healthcare
- Educational resources
- Discounts on products and services
- Savings and investments
- Maternity and paternity leave
- Generous time away
- Giving programs
- Networking opportunities
Overview
Join our Strategic Planning and Architecture (SPARC) team within Microsoft’s Azure Hardware Systems & Infrastructure (AHSI) organization and be a part of the organization behind Microsoft’s expanding Cloud Infrastructure and responsible for powering Microsoft’s “Intelligent Cloud” mission. We are seeking a masters/PhD student to join us in Cambridge winter/spring 2025 to work on model compression and optimization for LLMs, covering topics such as post training quantization and quantization aware training. You will be joining a welcoming and highly interdisciplinary team and work on creative and challenging problems during your internship.
Qualifications
Required/Minimum Qualifications:
- Be enrolled in Masters/PhD program in Computer Science/Machine Learning or related discipline
- Substantial experience quantization of LLMs, model compression
- Substantial knowledge in low-precision data type such as floating point, integer formats, block floats
Other Requirements:
- Cloud Background Check
Preferred/Additional Qualifications:
- PyTorch, Python, Hands-on experience in SW Tool development
- Outstanding communication skills
Responsibilities
- Research and develop quantization flow for LLM inference and training
- Design, implement and evaluate performance of quantized SOTA LLMs
- Write and present your findings in technical documents or presentations
Benefits/perks listed below may vary depending on the nature of your employment with Microsoft and the country where you work.
Industry leading healthcare
Educational resources
Discounts on products and services
Savings and investments
Maternity and paternity leave
Generous time away
Giving programs
Opportunities to network and connect
Set alerts for more jobs like Cambridge Internship in ML Model Optimization
Set alerts for new jobs by Microsoft
Set alerts for new Devops jobs in United Kingdom
Set alerts for new jobs in United Kingdom
Set alerts for Devops (Remote) jobs
