AI landscape of software engineering

• Дата: 2026
• Автори: Nesterenko, O.V., Fedorov, V.V., Yatsuk, P.P.
• Формат: Стаття
• Мова: Українська
• Опубліковано: PROBLEMS IN PROGRAMMING 2026
• Теми: artificial intelligence; SWEBOK; models and algorithms; maturity model; software; human machine interaction; UDC 004.4:004.8
• Онлайн доступ: https://pp.isofts.kiev.ua/index.php/ojs1/article/view/1021
• Назва журналу: Problems in programming

Artificial intelligence is becoming an integral part of modern software engineering, expanding the possibili ties of automation, optimization and support of technical solutions throughout the software life cycle. The article proposes a methodology for defining artificial intelligence tools for solving software engineering problems. The basis of this methodology is a system structuring based on the knowledge areas of the new version of the international manual SWEBOK V4 and a set of certain types of processes or activities (Top ics), which consist of knowledge areas. The main areas (six areas for software development) and auxiliary organizational areas (six areas that provide engineering and management of software development) were se lected for structuring. The application of such artificial intelligence models and algorithms in software engi neering as machine learning (ML), deep learning (DL), large language models (LLM), natural language pro cessing (NLP), Generative AI, graph algorithms, metaheuristics and optimization algorithms, expert systems and process analytics is considered. As a result, a general ontology of the AI landscape of software engi neering is built. A map of the application of AI in software engineering is considered in detail, which re flects the correspondence between SWEBOK knowledge areas, processes, AI technologies and tools. The proposed cartography demonstrates three important principles of integrating AI into software engineering: technological level; process level and instrumental level. This approach allows you to systematize the use of AI, avoid fragmented implementation of technologies, and also assess the level of intellectualization of de velopment processes. Separately, the article discusses the AI-Driven Software Engineering Maturity Model, which was created to help enterprises and organizations assess and improve their efforts to integrate AI and SE. The results of the study can be used to transform this symbiosis from an idea into a working paradigm, in particular, taking into account the reliability, explainability and ethical aspects of using AI. Problems in programming 2026; 2: 4-15