Abstract: Systems and methods are disclosed comprising techniques for personalization in code generation. A platform determines a user persona based on contextual information regarding a user associated with a natural language query. If a corresponding ontology entry is not found, wherein the code unit is identified based on the determined user persona, the platform applies an artificial intelligence model to generate a code unit by prompting the artificial intelligence model with the at least a portion of the natural language query and one or more of the user role information, the historical interaction pattern information, or the determined user persona. The generated code unit is executed against a data source to generate a result set. Responsive to detecting a feedback item in connection with the result set or the code unit, the platform enables incremental training of the artificial intelligence model. Code units can be modified based on additional feedback items.