"... do what thousands of highly qualified people in government agencies have so far not yet been able to do over decades of diligently trying..."
SpaceX have already done this with the Falcon rockets. But the plan here probably is to start the idea and put it out there, then the people who will actualize it (technically and financially) will join it when they see the potential, and when they see they are not alone. They will make sure the hard questions are asked and satisfactorily answered.
I'm bullish on the general idea.
You can use code to analyse data, help see patterns. The most important aspect though is that we are used to figuring out hard things, and know we must follow rules for something to work, and that giving up only means you don't get the thing done.