Agentic AI Engineer - RL

Responsibilities

• Design, implement, and train RL algorithms (PPO, A3C, DQN, SAC) for enterprise decision-making tasks
• Develop custom simulation environments to model business processes and operational workflows
• Build production-ready RL-driven agents capable of dynamic decision-making and task orchestration
• Integrate RL models with LLMs, knowledge bases, and external tools for agentic workflows
• Deploy RL agents on cloud and hybrid infrastructures such as AWS, GCP, or Azure
• Optimize training and inference pipelines using distributed computing frameworks like Ray RLlib
• Develop pipelines for evaluating agent performance, including robustness, reliability, and interpretability

Requirements

• 2–5 years of hands-on experience with Reinforcement Learning frameworks (Ray RLlib, Stable Baselines, PyTorch RL, or TensorFlow Agents)
• Strong programming skills in Python with proficiency in PyTorch or TensorFlow
• Deep understanding of RL algorithms including PPO, DQN, A3C, and Actor-Critic methods
• Experience with simulation environments such as Gymnasium, Isaac Gym, or Unity ML-Agents
• Exposure to cloud platforms (AWS, GCP, or Azure) and containerization (Docker, Kubernetes)
• Experience in reward modeling and optimization for real-world decision-making tasks

Preferred Qualifications

• Knowledge of multi-agent systems and collaborative RL
• Familiarity with LLMs and Reinforcement Learning with Human Feedback (RLHF)
• Experience with model optimization techniques like quantization, ONNX, or TensorRT

About the Company

XenonStack is a fast-growing Data and AI Foundry for Agentic Systems. We enable organizations to gain real-time intelligent business insights through our specialized platforms, including Akira AI for agentic systems, Vision AI, and NexaStack AI for inference infrastructure. Our mission is to accelerate the transition to AI + Human Intelligence by creating enterprise-ready AI agents that reason, adapt, and improve over time.