135K–300K USD / year
Research Engineer (Reinforcement Learning)
Machine learningRemote — North America
Published on 2026-08-18
These details were extracted automatically from the original listing. They may not be complete or fully up to date — it's worth checking the original job post.
About this role
LiveKit is seeking a Research Engineer specializing in reinforcement learning to enhance their voice and text AI models. The position involves building training environments, managing data pipelines, and ensuring model performance across various applications.
About the company
LiveKit provides open-source and cloud real-time media infrastructure for building voice, video, and AI agent experiences.
The team
R&D
Stack
PythonGPUs
What you'll do
- Build environments for model training
- Manage synthetic data pipelines
- Run and analyze training experiments
- Develop evaluation standards for model releases
- Select and adapt base models for specific tasks
- Improve model capabilities for voice and text agents
- Deploy models to production and enhance based on usage
What we're looking for
- Proficient in Python
- Experience in deploying models from data to production
- Focus on data quality and diversity
- Design against weak reward exploitation
- Familiarity with GPU limitations
- Ability to work collaboratively in a remote setting
Nice to have
- Experience in fine-tuning and reward design
- Familiarity with RL frameworks like TRL or OpenRLHF
- Experience with multi-GPU training techniques
- Knowledge of executing tool-using agents
- Proficiency in using open-weight models like Qwen or Llama
Benefits
- Competitive salary and equity package
- Health, dental, and vision benefits
- Flexible vacation policy