Week #5 #
In week 5, daily calls with the customer representative were scheduled. Starting next week, meetings are scheduled with metrics closely tracked using the computer vision model. The feedback received each day is immediately translated into tasks and implemented.
Our plan for collecting feedback is as follows:
- We form a report in the form of a document, where we describe the work done on 3 sides of the project: frontend, backend and ml. And also we describe how frontend and backend interact with ml part, as it is the most important for the customer. Example report:
- The representative reviews the report and generates feedback in one free-form telegram message as he/she sees fit. We get all other feedback in our daily meetings. Representative’s feedback from last week:
Translation: In the course of work to date, the team has implemented GUI templates, conducted research on the subject area, worked out the architecture, selected options for implementing the neural network and started its implementation . The following work has been done: GUI and back end, data partitioning within the framework of DataSet preparation for neural network training, a demo example of the system operation was realized, neural network training is performed, problem areas related to video quality and inability to recognize license plates of vehicles were identified.
- We form a report in the form of a document, where we describe the work done on 3 sides of the project: frontend, backend and ml. And also we describe how frontend and backend interact with ml part, as it is the most important for the customer. Example report:
Feedback on usability was given at the design stage, functionality is discussed daily on calls.
Every clarification that appeared in the process of receiving feedback is immediately converted into a high-priority task, because if the representative pays attention to it, it is of particular importance. Such tasks are performed first, and we show the result of the work done at the next meeting.
The main pattern that we have deduced is that the neural network arouses the greatest interest in the customer, while we perform other parts of the project in order to provide a more convenient service for further testing.
Often feedback affects each part of the project a little bit at a time, so each group within our team has 1-2 tasks that can be performed in parallel (e.g. ml and backend).
At the moment our team is reaching the goal we set at the beginning of the course and is moving along the plan. The development of a plan for further work on the customer’s task will be discussed afterwards, as it will strongly depend on the willingness to continue working of each individual team member.
Feedback
Collecting and documenting feedback
again you have only 1 feedback from Mark. now I’m not sure who is Mark. but is that enough? are you building the solution for 1 company? that’s it?beside, the documentation we ask her is to document the feedback not the project.
Documenting the project should be done by default.
Feedback analysis
i’m not sure that is this enough analysis. again you don’t have much feedbackRoadmap and enhancement
missingGrade: 2/5
Feedback by Moofiy