Senior Applied Scientist - London, United Kingdom - Microsoft
Description
Overview
From being able to log you in with face recognition, correct your gaze during a video chat to maintain eye contact, to the exciting possibilities of large generative models, are you itching to play a part in bringing applications of machine learning to millions of users?
The Microsoft Applied Sciences Group incubates disruptive technologies for Microsoft's next-gen hardware products and is working on several exciting projects that will shape how computers and other devices perceive the user and how users interact with computers. Operating as a startup within the company, this team works closely with several research and product teams to bring compelling new experiences to the market. A lot of these experiences will be powered by machine learning. As part of this team, you will have the unique opportunity to develop novel software solutions for different platforms that will revolutionize how people use computers in the future.
The team is growing, and we have an exciting opportunity for a talented software engineer to drive the development of web- and app-based state-of-the-art AI algorithms, bringing together the power of the cloud and ML-specific silicon on next generation devices and operating systems. You will have collaboration opportunities throughout the organization and will be building new stuff that really works and has millions of users.
Requirements
• Must be strong programmers with software engineering experience and excellent debugging skills.
• Must be passionate about incubating new ideas, solving problems, and building working systems.
• Must be self-motivated, proven collaborators, good communicators, attentive to details, and keen to learn.
• Practical experience working in a highly performant team in an academic or industrial research environment.
Qualifications
Qualifications
• Should hold an MSc or PhD in a STEM field
• or have a BSc in computer science or related fields with 3+ years of industry experience.
Required Experience:
• Minimum 3 years of C++ (especially C++11 and newer) and Python experience.
• Demonstrable input into machine learning, especially deep learning, driven product, familiar with frameworks like ONNX, pytorch.
Additional or Preferred Qualifications:
• Shipped products based on large ML models (SD, LLM).
• Familiar with Language models, transformers like BERT, GPT-2/GPT-3, Llama, OPT etc.
• Experience with prototyping new product functionality using LLMs
• Experience with CI/CD for MLOps, ideally AzureML
• Experience with model packaging and deployment.
• Experience in model quantization & optimization techniques such as GPTQ, LORA etc.
• Experience in creating APIs, SDKs and software packages for third parties, experience with WinRT, UWP, WinML and C# is a plus.
Responsibilities
Responsibilities
• Design and develop deployment strategies of ML models, such as Large Language Models (LLMs), from native code to packaging to web
• Develop task-specific plugins for LLMs
• Integrate LLMs and other large models to run on devices
• Integrate models, datasets, tests and training code into our landscape of products.
• Collaborate with engineering and product development teams.
• Contribute to a real-time system involving multiple components.
• Improve system quality through the design and implementation of new testing methodologies
• Integrate novel approaches into CI/CD system.