Research Scientist (Reinforcement Learning)

TL;DR

Research Scientist (Reinforcement Learning): Developing and scaling fundamental reinforcement learning algorithms for high-impact AI research with an accent on experimental rigor and large-scale model performance. Focus on implementing novel research hypotheses, conducting end-to-end experiments, and contributing to state-of-the-art developments in the field of AI.

Location: London, UK (Onsite)

Company

hirify.global is a world-leading research organization dedicated to pushing the boundaries of AI, developing transformative technologies like AlphaGo, AlphaZero, and Gemini.

What you will do

• Initiate and pursue novel research directions through testing and proposal of hypotheses.
• Implement and manage end-to-end experimental research projects.
• Build and improve research infrastructure at scale to support complex models.
• Analyze results, debug failure modes, and iterate on research implementations.
• Communicate research findings clearly through technical writeups and publications.
• Collaborate with interdisciplinary teams to empower researchers and scale experiments.

Requirements

• Research track record in reinforcement learning, including peer-reviewed publications.
• Strong implementation ability and experience with research codebases.
• Evidence of owning research experiments end-to-end.
• PhD in machine learning or equivalent practical experience.
• High agency, strong prioritization skills, and ability to take initiative.
• Excellent communication skills with a bias toward transparency and clarity.

Nice to have

• Experience with sequence models, post-training, or preference-based learning.
• Proficiency in modern research stacks such as JAX/Flax or PyTorch.
• Strong experimental judgment regarding baselines and ablations.

Culture & Benefits

• Collaborative environment with a tight-knit, world-class research team.
• Commitment to diversity, equity, and inclusion in the workplace.
• Strong emphasis on continuous learning and professional development.
• Opportunity to influence cutting-edge AI breakthroughs and product impact.