Agentic Development Framework
Investigating structured processes for reliable, repeatable development with AI coding agents.
Research directions, experiments, methods and scientific publications. Explore what we investigate, how we evaluate it and what we learn.
Research entries describe topics under investigation. Scientific publications link to their source and verified bibliographic details.
Investigating structured processes for reliable, repeatable development with AI coding agents.
Specialization, coordination and verification in workflows involving multiple AI agents.
A systematic view of correctness, maintainability, effort and autonomy.
Repository structures that give agents the context and boundaries to work effectively.
Investigating reusable ways to plan, implement, hand off and verify agent work.
Connecting task specifications, acceptance criteria, review and human intervention.
Vendor-neutral investigation of tools, models, environments and orchestration.
A systematic literature review of AI-driven assistants in scaled agile development, examining adoption challenges and potential support.
A controlled experiment on generative AI for agile effort estimation, examining estimation accuracy and support for planning.
A review and evaluation of intelligent code completion assistants, including functionality and developer usability.
An empirical comparison of self-admitted and tool-measured technical debt in 14 production software products.
A controlled experiment on AI-assisted decomposition of user stories into development tasks using GitLab Duo.
A systematic mapping review of testing challenges and approaches for AI-based systems, including gaps in the available evidence.