Applied Scientist

Microsoft•Redmond, WA

About The Position

Frontier Tuning is Microsoft’s AI customization platform, enabling enterprises to adapt foundation models to their workflows, domains, and data while preserving security, privacy, and reliability at scale. The Applied Scientist will help advance enterprise model customization across Microsoft 365 and beyond, combining algorithmic innovation with the development of scalable infrastructure and tooling for training, steering, evaluating, and securely deploying enterprise-ready AI systems. The role may involve scalable training systems for RLHF/RLAIF and other post-training pipelines; transformer architecture design or efficient adaptation methods such as LoRA-style techniques; inference-time steering, controllability, or alignment; foundation-model debugging, evaluation, and development tooling; and operating and maintaining AI production systems. The scientist will implement and evaluate adaptation approaches under enterprise deployment constraints, including latency, safety, privacy, policy compliance, and compute efficiency. They will partner with research and engineering teams to turn product or customer requirements into scalable solutions, take technical work from concept through prototype, and document approaches and share best practices for post-training and secure deployment.

Requirements

Bachelor’s degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or a related field and 2+ years of related experience, or a master’s degree in one of these fields and 1+ year of related experience, or a doctorate in one of these fields, or equivalent experience.
Ability to meet Microsoft, customer, and/or government security screening requirements, including passing the Microsoft Cloud Background Check upon hire or transfer and every two years thereafter.

Nice To Haves

Bachelor’s degree in a listed or related field and 5+ years of related experience, or a master’s degree and 3+ years, or a doctorate and 1+ year, or equivalent experience.
1+ year of experience creating publications, such as patents or peer-reviewed academic papers.
Experience contributing to open-source systems or production deployments involving foundation-model training or adaptation.
Experience in one or more of: reinforcement learning or post-training methods; distributed or large-scale ML training systems; AI-system evaluation or benchmarking; tool use, planning, or agentic model behaviors; or deploying AI solutions in enterprise or customer environments.
1+ year of experience building scalable ML systems or pipelines for training, adapting, or deploying AI models.
2+ years of experience with Python and machine-learning frameworks such as PyTorch or equivalent.
Experience with scalable RLHF/RLAIF or other post-training systems, transformer architecture design, efficient adaptation, inference-time steering or alignment, foundation-model tooling, or operating AI production systems.

Responsibilities

Implement and evaluate model adaptation approaches under real-world enterprise constraints, including latency, safety, privacy, policy compliance, and compute efficiency.
Partner with research and engineering teams to translate product or customer requirements into scalable model adaptation solutions.
Drive technical work from concept to prototype, delivering methods, system components, or empirical insights that advance enterprise model customization.
Document approaches and share best practices to improve organizational capabilities in post-training and secure foundation-model deployment.
Contribute to work involving efficient hosting of large language models and parameter-efficient adapters; adaptation techniques; inference-time steering or controllability; and tooling, harnesses, evaluation, debugging, or model development.