Biotech needs a serious kick in the pants. It has always been remarkable to me just how little automation is used in the field of molecular biology. There is a huge amount of work that involves humans doing repetitive tasks and then waiting for something to happen. This increases cycle times, limits the number of hypotheses you can test, and is rate-limiting for the whole field. You already have the problem that you are working on living things which need time to grow, reproduce, eat, etc...

Not saying any of it is easy to automate by a long shot. But that kind of innovation is going to be what is needed to propel the field forward.