The 2022 NeurIPS CityLearn Challenge focused on developing reinforcement learning agents for energy management in buildings. Our solution used a combination of model-based and model-free reinforcement learning techniques to optimize energy usage while minimizing costs.