Reinforcement Learning for Frontier AI Models
Microsoft
In this role you contribute to the large-scale reinforcement learning efforts at Microsoft AI, shaping models through data collection, evaluation, and advanced RL techniques. You will work within an interdisciplinary team to improve pre-trained models and their real-world performance, focusing on reasoning, instruction following, and agentic capabilities. The role emphasizes end-to-end ownership, rapid experimentation, and collaboration with cross-functional partners. This is a chance to impact billions of users by advancing state-of-the-art RL for next-generation AI systems.
Responsibilities- Contribute to data collection, evaluation, and RL methods for models
- Design hypotheses and experiment plans to rapidly iterate on model performance
- Demonstrate and uphold culture and values within the team
- Bachelor's degree in Computer Science, Machine Learning, Mathematics, or related field or equivalent experience
- 4+ years of software engineering experience
- Experience with reinforcement learning and LLMs
- strong collaboration and communication skills
- high attention to detail and craftsmanship
- end-to-end ownership of projects
- reinforcement learning
- RL environment creation
- using real-world data and feedback to improve models
Reference: WJ-747_31294675 · Posted 14 Nov 2026, 11:40 PM