Reinforcement learning is a fascinating field—agents learn to solve complex problems through trial and error, guided only by reward signals. This project involves implementing a reinforcement learning agent in a simplified Pong environment using Curriculum Learning and Proximal Policy Optimization (PPO). Curriculum Learning breaks down the training process into stages of increasing difficulty, allowing the agent to master each stage before advancing. The PPO algorithm uses a clipped objective to increase the stability of updates across stages.