Reinforcement Learning Engineer in United States at Jobgether

This position is posted by Jobgether on behalf of a partner company. We are currently looking for a Reinforcement Learning Engineer in the United States.

This role focuses on designing, training, and deploying advanced reinforcement learning systems that solve complex sequential decision-making problems where traditional supervised learning approaches are insufficient. You will work on building intelligent agents that learn through interaction, with applications spanning simulation environments and real-world production systems. The position blends deep research in modern RL methods with hands-on engineering to ensure models are scalable, stable, and safe in production. You will contribute to shaping reward systems, training infrastructure, and evaluation frameworks that directly influence model behavior. The environment is highly technical and research-driven, requiring close collaboration with applied scientists and product teams. This is a high-impact role where your work will transition cutting-edge RL techniques into production-ready systems. You will help define how intelligent agents are trained, evaluated, and continuously improved at scale.

• Design and implement reinforcement learning systems for sequential decision-making problems across simulated and real-world environments.
• Develop and maintain high-fidelity simulation environments to support scalable agent training and experimentation.
• Implement and evaluate RL algorithms including policy gradient, actor-critic, off-policy, and offline reinforcement learning methods.
• Engineer reward functions and shaping strategies that align model behavior with performance, safety, and business objectives.
• Apply offline RL and imitation learning techniques in environments where exploration is constrained or unsafe.
• Utilize RLHF, DPO, and related approaches to fine-tune large-scale models where applicable.
• Build distributed training infrastructure for RL, including experience replay systems and scalable data pipelines.
• Improve training stability and sample efficiency through algorithmic optimization and systems-level enhancements.
• Design rigorous evaluation frameworks, including adversarial testing and out-of-distribution validation.
• Implement safety mechanisms such as constraints, guardrails, and human-in-the-loop oversight systems.
• Collaborate with applied scientists and product teams to identify and prioritize high-impact RL applications.
• Monitor production models for drift, performance degradation, and unintended behaviors, building observability tools and alerting systems.
• Document methodologies, system design, and operational processes for long-term maintainability and knowledge sharing.
• Stay current with reinforcement learning research and translate novel techniques into production-ready solutions.

• Master’s or PhD in Computer Science, Machine Learning, or equivalent practical experience.
• 6+ years of combined experience in reinforcement learning research and engineering.
• Strong programming skills in Python and deep learning frameworks such as PyTorch or TensorFlow.
• Hands-on experience with RL libraries or custom RL training stacks.
• Solid understanding of probability, optimization, and reinforcement learning theory.
• Experience designing reward functions in complex or high-dimensional environments.
• Familiarity with simulation environments and large-scale training pipelines.
• Experience training neural policies on GPU-based distributed systems.
• Strong debugging, experimentation, and analytical skills.
• Excellent communication skills with a track record of shipping or publishing impactful RL work.

• Competitive compensation aligned with experience, typically in the $100,000–$150,000 range.
• Fully remote position within the United States.
• Long-term, multi-year engineering engagement.
• Direct W2 employment with full benefits package.
• Opportunity to work on cutting-edge reinforcement learning systems in production environments.
• Exposure to large-scale AI training infrastructure and advanced model optimization techniques.
• Strong focus on research-to-production impact in a high-growth technical environment.
• Collaborative, research-driven engineering culture.