International Workshop on AI for technical debt management in agile software development (AI4TD)XP 2026

In modern agile software development, the ability to obtain fast, actionable feedback on the accumulation of Technical Debt (TD) and its associated risks is paramount for developers and managers. The rapid adoption of AI in software engineering brings a dual reality for TD: Fast paced AI-generated code can introduce a vast amount of TD, while new AI-powered approaches also offer capabilities to detect, quantify, and help remediate debt. Agile teams in industry face pressure to deliver faster and incorporate AI-assisted tools into their workflow, which changes how debt accumulates, is perceived, is managed within backlogs, sprint planning, and release pipelines.

At the same time, software processes are becoming increasingly data-driven, leveraging data at scale in innovative ways. This creates an opportunity: using AI to enable data-driven TD management in agile environments. The socio-technical implications of such integration on how TD is managed can be vast, reducing dependencies but also the opportunities for learning crucial architectural constraints.

This workshop offers a platform for researchers and practitioners to share ideas, experiences, and future visions on applying AI for TD management in agile development.

Call for Submissions

OBJECTIVES AND TOPICS

The guiding theme of the AI4TD workshop is:

What implications does artificial intelligence (AI) have in the accumulation and management of technical debt in Agile development? AI4TD provides a venue for researchers and practitioners to exchange and discuss trending views, ideas, state-of-the-art, work in progress, and scientific results highlighting aspects of AI and TD. This is the first proposed workshop dedicated to the discussion of how Artificial Intelligence and Technical Debt are interrelated, how to exploit opportunities and highlight challenges, especially dedicated to Agile software development. Being the first of its kind, and being AI and Technical Debt a new topic of interest, we expect quite some novel contributions to the workshop. Topics include but are not limited to:

• AI generating new types of technical debt;

• Quantification of AI-generated technical debt;

• AI generating code with less technical debt;

• AI suggesting refactoring for technical debt;

• AI using context engineering to better understand technical debt;

• AI suggesting software architectures-aware solutions;

• Using AI to suggest a prioritization of TD;

• AI addressing TD in modern software architectures like microservices;

• AI addressing security technical debt;

• The implications of using AI on the teams’ skills to manage TD

• The implications of using vibe coding to generate code on architectural technical debt;

• The implications of using AI to refactor code-level TD on architectural technical debt;

• Tools and methods to integrate AI into the management of TD;

• Best practices to use AI for technical debt management;

• Trade-off using AI, keeping important technical debt at bay, and social implications

• Experiences using AI for technical debt education

SUBMISSION GUIDELINES

We invite the following types of contributions:

• Paper-oriented contributions (9 pages max), which concentrate on presenting long research papers in detail and are followed by discussions.

• Discussion-oriented contributions (6 pages max), which – facilitated by brief presentations of short research papers – emphasize longer discussions.

Discussions will be facilitated by the workshop chairs by preparing a set of guiding questions for the presenters.

• Industrial experience reports (6 pages max), which address problems and lessons learned from practice and are accompanied by a summary paper.

• Tool or method demo (4 pages max), which provides an interactive session to report lessons learned on using AI for technical debt management or challenges with AI-generated code.

• Lightning talks (2 pages max), which provides a teaser for a new challenge or opportunity to be discussed among the participants.

For submission, please use the following guidelines and template: https://www.springer.com/gp/computer-science/lncs/conference-proceedings-gui

Important dates for submissions:

• Workshop paper submission deadline (not for “tool and method” and lightning talks): January 23, 2026

• “Tool and method” and “lightning talks” submission: March 4, 2026

• Workshop paper selection by workshop chairs, notification to authors: February 20, 2026

• Workshop selection for “tool and method” and lightning talks by workshop chairs, notification to authors: March 9, 2026

• Workshop day at XP2025: April 8, 2026

• Camera-ready workshop papers and copyright forms uploaded (not for “tool and method” and lightning talks): April 24, 2026

Workshop proceedings of accepted long and short papers (not tools and demo and lightning talks) will appear in the digital library. For rejected papers, it will still be possible for authors to be invited to give a lightning talk or to showcase a tool demo.

PAPER SELECTION PROCEDURE

AI4TD will follow a single-blind review process. Each submission will receive at least three reviews from the members of the program committee. Following the review submission, there will be an online discussion period. The organizing committee together will make the final decision on acceptance. For “tool and method” and “lightning talks” a more informal process will be used to evaluate the submissions with the aim of being as inclusive as possible. These submissions will not be included in the proceedings.

TENTATIVE SPECIAL ISSUE OF JOURNAL

If we receive enough high-quality submissions, we plan to invite the best workshop papers to extend their work in a special issue of a well-established journal in the field of software engineering. More information will be shared after the 23rd of January.