This list contains only the countries for which job offers have been published in the selected language (e.g., in the French version, only job offers written in French are displayed, and in the English version, only those in English).
Frontier Tuning aims to fine-tune frontier LLMs on enterprise data, enabling task-specific agents and solutions. We are a small, nimble team that is advancing the state of the art of models in M365 Copilot. Come join our team and help transform the LLM experience in the enterprise.
Job Responsibility
Train and deploy Language Models adapted to specific industry needs
Create and adapt novel training and fine-tuning algorithms for language models with special focus on reinforcement learning for long-horizon dynamic workflow
Research innovation and scholarly dissemination: conceive and execute research projects that advance training methodologies
write and submit peer-reviewed papers or preprints
and present work at conferences
Drive end-to-end translation of research into product capabilities, leading projects from ideation and prototyping through production integration and measurable customer impact
Requirements
Bachelor's Degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or related field AND 2+ years related experience
OR Master's Degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or related field AND 1+ year(s) related experience
OR Doctorate in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or related field
OR equivalent experience
1+ years of experience training/fine tuning AI/ML models, preferably large language models
1+ years of experience building Generative AI pipelines, e.g. with RAG
1+ years of experience with Python and/or PyTorch
Ability to meet Microsoft, customer and/or government security screening requirements
Microsoft Cloud Background Check
Nice to have
Experience with multi-agent training in dynamic harness
Experience training or contributing to the development of very large-scale language models (e.g., 100B+ to trillion-parameter models), including distributed training, async RL, and long sequence handling