Hierarchichal RL framework for intraday optimal trade execution. Implements a two-layer architecture: Strategic PPO agent selects execution pace, Tactical DQN agent optimises order slicing and timing All experiments conducted on simulated market data with reproducible seeds for controlled benchmarking. Outperforms traditional algorithms.