The digitalization of education facilitates access to information; however, beginners still face challenges in understanding programming languages such as PHP. This study aims to develop a chatbot based on Natural Language Processing (NLP) using the Sentence-BERT model (all-MiniLM-L6-v2) to understand user questions in natural language contextually. The research follows a prototyping development method, consisting of several stages: needs identification to determine relevant features for users; interface design to create an intuitive and user-friendly layout; web-based system implementation to realize system functions; and testing using the black-box method to ensure each feature works as specified, along with usability evaluation using the System Usability Scale (SUS) to assess user comfort and ease of use. The result is a chatbot application capable of matching user questions with a Q&A database using semantic similarity. All testing scenarios ran as expected. The SUS evaluation yielded a score of 89.58, indicating a very high level of user satisfaction. This research demonstrates that the integration of NLP and BERT can enhance the effectiveness and convenience of independent programming learning and has the potential to be applied to other educational platforms.