
Applied Research - Forward-Deployed RL
AI
Prime Intellect is hiring an Applied Research engineer for Forward-Deployed RL to own applied RL programs from ambiguous customer problems through deployed improvement loops. The role involves building environments, evaluations, and reward signals, running post-training experiments, and deploying results with customer teams. Requires strong Python, hands-on model-training or post-training experience, and strong experimental judgment.
What you'll do
- Own applied RL programs from first ambiguous customer problem through a deployed improvement loop
- Understand the workflow and business stakes and decide what can be measured and learned
- Build the environment and evaluation
What you need
Must-have
PythonReinforcement LearningPost-trainingDistillationAgent systemsModel trainingEvaluationTool use
Soft skills
Experimental judgmentComfort with ambiguityCommunicationStakeholder managementWillingness to change direction
Applied ResearchReinforcement LearningPost-trainingAI InfrastructureAgent SystemsPythonOpen SourceFull-timeLondonRemote