Operating Systems (OS) is one of the most fundamental blocks of any computer. This basic, though crucial layer acts as the middlemen between the user and the processor performing thousands of tasks for better operational efficiency. One of the most important tasks that OS handles is how each process will be executed within the Central Processing Unit (CPU). For any normal user there might be many processes that the CPU needs to handle to meet the user’s computational requirement. Thus, it is extremely crucial for any OS to handle these tasks in a specific method that meets the user’s needs at the same time preserve resources for the system to function properly. Algorithms like First-Come-First-Serve (FCFS), Shortest-Job-First (SJF), Priority-Scheduling, Round-Robin (RR), etc. are extremely prevalent in our modern systems. However, as the complexity and organization of computing systems increases dynamic scheduling algorithms becomes a necessity. This research explores a competitive way of integrating Machine Learning (ML) techniques to Genetic-Algorithm (GA) for efficient CPU scheduling. The proposed method incorporates ML-predicted best-fit functions into the GA framework. Comparative analysis with traditional scheduling techniques, including FCFS, SJF, and Round Robin, demonstrates superior performance. Overall efficiency is evaluated using a novel metric that combines throughput and waiting time. Experimental results highlight the effectiveness of the proposed approach in achieving optimal scheduling outcomes.