From an annotation perspective: the path tool and the assisted annotation tool are most useful for our needs, and earlier bugs related to these tools appear to be fixed now. The zoom feature is also more user-friendly than earlier in beta testing. One concern earlier in annotation experience was being able to return to already annotated images to fix a mistake or to review: the inclusion of the Work history button seems to meet this need to view what images have been annotated and return to previously annotated images in the batch. The addition of an a confirmation pop-up message before submitting an annotated image is also helpful and helps reduce user error. During this beta phase, bug reporting on the Innotescus website and email communication about updates and fixes were helpful and efficient. Form the customer support perspective: we do appreciate the added feature that allows you to change object class via right clicking. Overall, the tool is working well for our needs. In addition, the customer support process has been smooth and efficient to the point that several of the requests for new features have been dealt with efficiently, which builds trust on the customer side. Nevertheless, we are sure that there is room to improve. The annotation process is often very inconvenient, especially when a coordinated process is required between several individuals on a team. Fortunately, Innotescus has offered us a set of tools that has streamlined our work in an unbeatable way. Before using the Innotescus platform, we had tried to use open source tools that unfortunately hindered the annotation process, thus seriously affecting the team's productivity level. We also tried a couple of commercial tools, however the learning process was torturous. Fortunately for us, the Innotescus platform brought with it several advantages, including an extremely simple and compressed documentation that conveys exactly what end users need to know when using a tool. This has a direct significant impact on the learning curve that is reflected in the level of productivity of the team. In addition, the various modalities for annotation tasks help to simplify and create a pipeline that is then part of a sophisticated machine learning scheme such as multi-modal. This is largely reflected through the annotations export process in both COCO and Pascal VOC formats. Finally, regarding the management process at the admin level, it is extremely simple and requires very little effort to be able to start assigning roles, create projects, tasks and statistically evaluate the different characteristics of the datasets with which you want to work within a team. . This in turn allows to have a deeper and clearer vision of the characteristics of the data to be processed, and thereby choose techniques and algorithms that better fit those conditions.