Totally agree. In one of our projects (around 200K Python LoC) we had 90% CodCov, but we frequently found some bugs.
Recently we started to heavily use mutation testing and fuzzing to find edge cases on parts of the code that were already "covered" according to the CodCov report. It was definitely worth it.
I highly recommend investing in mutation and fuzzy testing.
I also run a company [1] that primarily recruits low-income Brazilians from remote locations and teach them how to code and create data-related products. The company itself is for profit in the long run, but the first priority is to make our members grow (financially and psychologically), then generate profit for the founders.
During the entire process we focus 40% of their time to study/learn about soft skills (become a better person/human), 20% for hard skills (become a better professional) and 30% for producing code and create products, we pay 100% for all the courses and books they use to lean.
Every quarter we review and increase their monthly income based solely on their level of soft skills, not hard skills. So if they become more organized, have better communication, etc. they earn more. People can start as low as R$ 500,00 per month and grow up to R$ 25,000,00 per month.
In the end, our goal is to make each of them more (or totally?) lucid/self-aware (in the context of psychology) and financially independent (more than R$ 1 million on their bank account) in the maximum period of 10 years [2].
We have been doing it for the past 5 years, fully remote, and it's currently very close to becoming a sustainable process, as some of our current members are very lucid/self-aware and getting close (2-3 years away) to become financially independent.
For Python we use mutmut [1].
[1] - https://pypi.org/project/mutmut/