There's a pretty big project which will be checked 2 testers one Junior with little experience, one with extensive experience in this project.
There are a few problems
1. Regrese almost fails
2. The amount of bugs/resolve from sprint to sprint every time a lot
3. No time to write test cases and maintain documentation
4. Testing feature almost always results in a bunch of bugs, and then returned kom resolves which prevents test features
- On what technologies the app?
- What is the sprint length?
- How you built the process of code writing and integration?
- How many developers?
- What roles besides developers and testers still have on the project?
How do you do architecture, if you can't add functionality without blowing a significant part of the whole system?
The idea of code before mergem must be checked to test the Assembly. Maybe you need diopter if you don't.
During inspection (testing) can be written in this script acceptance test. So you have to slowly appear automated tests.
I would prioritized acceptance tests over unit tests in this case. Automated acceptance tests directly unload testers who have, according to the size of the project, is not enough. The use of unit tests is only visible in the long term.
The developers also needs to be tested. The one who writes feature X check feature Y, and Vice versa. It is impossible to check their own work. We need a culture of quality and open communication to developers is not suppressed seen the problems, and openly wrote the ticket.
Go into the next sprint with bugs from the previous - do not ice. You can make one sprint - adding functionality, the second sprint stabilization.