Principal Software Engineer

Key details

Full job description

The team delivers efficient GPU techniques for Microsoft's online advertising platforms, focusing on leveraging large language models (LLMs) and deep learning to improve efficiency and monetization. The Principal Software Engineer will work on end-to-end stack development emphasizing GPU use for LLMs. Responsibilities include designing and implementing inferencing capabilities for deep learning models, building extensible frameworks, optimizing inference performance on hardware/software stacks, collaborating to enhance model deployment, developing AI lifecycle tools, and fostering an inclusive team environment. Required qualifications include a Bachelor's degree in Computer Science or related field with 6+ years of engineering experience coding in C, C++, C#, Java, JavaScript, or Python, and passing Microsoft Cloud background checks. Preferred qualifications include a Master's degree or equivalent experience with 8+ to 12+ years of experience, 2+ years with deep learning frameworks (PyTorch, OnnxRuntime, Tensorflow, vLLM, TensorRT-LLM), experience in system design and MLOps, GPU architecture and kernel programming expertise, Docker and Kubernetes knowledge, and technical leadership skills. The role is full-time, on-site in Redmond, WA. Salary ranges from $142,800 to $274,800 annually in the U.S., higher in select metro areas.