Assessing students’ programming assignments, which measure their understanding and proficiency with coding concepts, is among the most important aspects of Computer Science education. The rising number of students enrolling in programming courses has historically made this process more tedious, subjective, and time-consuming. The Development of automated programming assessment systems (APAS), which offer quick, consistent, fair assessments, is one way to handle these problems. APAS enhances the learning environment by simplifying the grading system and supporting fairness and transparency, freeing instructors to devote more time to teaching. Seeking to enlighten educators and researchers about the state of APAS and pinpoint areas for further research, this systematic review has been done to evaluate the present approaches and technological breakthroughs in APAS fully. The main focus is standard APAS techniques and their effectiveness in offering timely, helpful feedback in learning environments. The findings indicate that the primary grading metrics of APAS include correctness, readability, maintainability, and documentation, while its testing strategies are categorized into dynamic testing and static analysis. It has also appeared that most APAS possess feedback features. However, they face limitations in providing comprehensive feedback essential for improving students’ programming skills. Future research should focus on improving the feedback systems by providing timely and personalized student feedback, which may help improve learning outcomes and enhance programming skills. Furthermore, APAS could also benefit from integrating some artificial intelligence and machine learning techniques into its testing strategies.