What is it about?
Architecting software-intensive systems can be a complex process - sketching a blueprint - that guides software implementation and evaluation. Software Development Bots (DevBots) trained on large language models can help synergise architects' knowledge with artificially intelligent decision support to enable human-bot collaborative software architecting. This research teams up a novice software architect and ChatGPT as a DevBot to architect a software system.
Featured Image
Photo by Mohamed Nohassi on Unsplash
Why is it important?
To explore and illustrate how large language model (LLM) based conversational bots like ChatGPT can undertake software architecting activities - enabling automation and providing decision support - to design, develop, test, and maintain software systems effectively and efficiently.
Perspectives
This research investigates the potential and perils of ChatGPT as a software development bot (DevBot) that can assist and empower the role of an architect who leads the process of architecting to develop software-intensive systems. The research needs empirical validations, grounded in evidence and experimentation, to objectively assess factors like enhancing engineers’ productivity, architecting process optimization, and assisting novice developers and designers to engineer complex and emergent classes of software effectively with ChatGPT.
Aakash Ahmad
Lancaster University
Read the Original
This page is a summary of: Towards Human-Bot Collaborative Software Architecting with ChatGPT, June 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3593434.3593468.
You can read the full text:
Contributors
The following have contributed to this page
