About the job
Poolside is dedicated to advancing Artificial General Intelligence (AGI), with a focus on innovation and engineering that drives economic and scientific progress. The company brings together experts in research and software development, all working toward high-quality systems.
As a remote-first team, Poolside’s staff are based across Europe and North America. Team members meet in person for three days each month, with longer offsite sessions twice a year. This structure supports collaboration among people with both research and engineering backgrounds.
Role overview
The Reinforcement Learning Infrastructure Engineer joins the reinforcement learning team to improve reasoning and coding capabilities in Large Language Models (LLMs). The role covers the full process: researching new algorithms, designing and scaling RL environments, and implementing solutions across the stack. Work is supported by access to thousands of GPUs.
Core mission
Build and scale infrastructure for reliable, efficient LLM training using advanced reinforcement learning methods.
Key responsibilities
- Stay current on research and developments in LLMs, reinforcement learning, and code generation.
- Develop strategies for fine-tuning training and inference, ensuring integration throughout the development process.
