That's what happened to me yesterday. I missed almost the whole thing as my calendar was booked solid. Only got to catch the tail end. Not a good look to put someone on hold to trade ETH.
Bot co-development effort? I've improved a lot at the micro movements since I bought my first crypto. I'm ashamed I got caught panic selling twice early on before smartening up. The longer time horizon movements are trickier, but I'm starting to see them earlier too, and have been developing some rules which I'm testing. I don't think I can spot signs of the complete turnarounds until they happen though.
I have been thinking about for a while, and eventually decided to divide the bot into two different projects. One would be the development of an app that takes care of executing buy and sell orders, and the other would be the development of the for app that makes the decisions. The reason I decided on separating the two is that they require different skills. The first one would require the knowledge of securely working with exchange APIs, while the latter is mostly data analysis.
I am not a pro trader, so I obviously lack a lot of knowledge, but the obvious shortcoming that I see is that I am not consistent in the decisions that I make, which is partly because there are too many parameters that I need to check to make a decision, and in a market this volatile it is easy to miss some of the parameters or haste to misread them to confirm an emotional decision. So, for now, I'd be content to just have the latter half of the app: essentially an app which reads a stream of data from the exchange's API, uses the predetermined algorithms to decide upon a buy/sell signal, and alert me by email or something when it detects an entry point (then I'll put the order manually).
At first, the app will not be intelligent, it will just check the data against some predetermined criteria to come up with buy/sell signals; however, later on, with the help of machine learning or at least some simple statistical analysis, it can be used to develop some adaptive/tuned decision making techniques.
Edit: My to go programming language is typescript (I can use it for both backend development with node and frontend development with Angular), but TradingView has a custom programming language Pine which is used for working with its chart, and it seems sophisticated enough to be used for an initial effort. I haven't really looked into it in any detail though