Skip to content

Retrospective‐Sprint 3

Isabella Acosta Pareja edited this page Apr 26, 2025 · 6 revisions

During this sprint, we focus on presenting the progress in the implementation of artificial intelligence within the project. We show how AI has been integrated into key functionalities, explaining the development process and the results obtained so far.

What should we continue to do (best practices)?

It is important to continue working on the two requirements that were left in progress during this sprint: the history of recommendations and the availability of books. Improving them will allow us to move towards a more complete and functional solution.

What should we start doing (process improvements)?

We need to start cleaning up the code more systematically. Maintaining an orderly and well-structured code will facilitate teamwork and reduce potential errors in the future.

What should we stop doing (problems and bottlenecks)?

We should avoid working in isolation. It is essential to maintain constant communication between all team members to keep abreast of changes and progress in the project. This will help us to prevent misunderstandings and improve collaboration.

Clone this wiki locally