Manually building and visualising a decision tree

I am currently working on a decision tree for predicting the average of a binary outcome with a small number of categorical features (think predicting the survival rate in the titanic dataset). I am aware of different R and Python libraries that allow to optimally calibrate decision trees on historical data. However, I’d like a more manual, interactive -ideally without code, ideally free- approach to build simple decision trees.

I am fully aware that this would not yield optimal results in many senses. However I have many practical constraints that would make that an interesting solution. This constraints include, but are not limited to : (for the design and the usage of the model) variables playing a key role in the industrial process should ideally be put at the top, need to involve non-coding people in the design of the tree, need to use limited and technically meaningfull splits so that user can print the tree and go trough it in less than 10 sec. and might even memorize parts of it). And, maybe most importantly in terms of modelling, the need to go further than calibration on past historical data and include prospective expert-based grouping of behavior in the model, as the current crisis is likely to change a lot of things.

So what tool, ideally free, ideally with a graphical user interface, and compatible with some well established DS / corporate langage (R/Python or Excel…) might be used for manually building some small non-binary tree / change splits / visualise the tree and outcomes ?