Senior Applied Scientist: Microsoft

The Azure Language Artificial Intelligence (AI) team is seeking a Senior Applied Scientist to help accelerate the development of agentic AI systems. This role focuses on advancing the scientific methods and scalable infrastructure needed for training agentic models to achieve frontier-level performance. The work involves contributions to large language models (LLMs), small language models (SLMs), and advanced data synthesis to build reliable, enterprise-grade agentic workflows. The ideal candidate is a self-driven, curious, and adaptable problem-solver who can write efficient code and document their findings effectively.

Important Notes:

Travel requirement is 0-25%.

This position is subject to a Microsoft Cloud Background Check upon hire and every two years thereafter.

Responsibilities:

Advance scientific methods and scalable infrastructure for training agentic models.

Contribute to the development of large language models (LLMs), small language models (SLMs), and agentic models.

Utilize and advance data synthesis techniques with both proprietary and open-source frameworks.

Build reliable, enterprise-grade agentic workflows.

Write efficient code, document findings, and demonstrate a track record of impact in the field.

Required Qualifications:

A Bachelor's Degree in a relevant field (Statistics, Computer Science, Engineering, etc.) and 4+ years of experience, OR a Master's Degree and 3+ years of experience, OR a Doctorate and 1+ year of experience, OR equivalent experience.

2+ years of experience developing GenAI-powered products through multiple stages of the product lifecycle.

2+ years of experience in data curation and synthesis for model optimization.

2+ years of hands-on coding experience in Python and frameworks like PyTorch.

Experience with both proprietary and open-source frameworks.

Preferred Qualifications:

A Doctorate in a relevant field and 2+ years of experience OR a Master's Degree and 3+ years of experience.

Extensive experience with fine-tuning LLMs, SLMs, and agentic models.

Proven track record of impactful research through published work or real-world product deployments.

A startup-style mindset: agile, solution-oriented, and self-driven.

Ability to collaborate across interdisciplinary teams and communicate complex ideas clearly.