The domain of card games, characterized by their inherent complexity, uncertainty, and strategic depth, presents a fertile ground for exploring the capabilities and advancing the frontiers of Reinforcement learning. Our research employs a combination of reinforcement learning (RL) and Deep reinforcement learning (DRL) techniques, including Deep Q-Networks (DQN), Counterfactual Regret Minimization, Deep Monte Carlo (DMC), and Neural Fictitious Self-Play (NFSP) to navigate the intricate strategic landscapes of the BlackJack and Poker games. The results showcase the algorithms' capability to achieve superhuman performance in complex card games and provide insights into the mechanisms of strategic learning and decision-making processes.