It is only a partial solution. You need the pip and virtualenv packages along with a private PyPI repository. Binary wheels make it so that you don't have to clutter any environments except for build servers with compilers and other build tools. There's still the problem that Linux wheels don't differentiate on distribution, so you will hit problems if you build on one distribution and try to deploy on a very different one.
Use PyPICloud or DevPI to host your built binaries. Wheels have another advantage that you don't install any dev tools in your production environments (no need for compiler tools). The only annoying thing is that there isn't any way to distinguish between different Linux distributions, so hopefully all your servers are standardized.
You can get a government contract. Normally this would not be available to you even if you could handle the work because of how contracting for the government works. What I will find interesting is if they solve the contract vehicle/payments problem. That is, the U.S. government isn't great about prompt payment. For a large contractor, they can afford to wait 3-6 months between payments. However for the small vendors being targeted with this, that kind of pay delay may kill their business.
Don't underestimate the value of experience. While failing must certainly have hurt, you have actually learned more than if everything had worked out. Whatever you do, be sure to take a bit of time to regroup before you jump into another high intensity activity like your next startup or someone else's software project with a ridiculous timeline. That way you will recover quicker.
The wrong people are working on this problem. The only ones who think about it are typically programmers who have made their peace with the machines as they are today. The interface of code seems simple and logical with little reason to try to improve on it. It would take a team of artists, musicians, human factors engineers, ethnographers and some clever computer scientists to do it. Such an enterprise would be high risk and very difficult to fund because of it.
I bet you now know a lot about parsing and the relational schemas and managing models now. You'll hit a problem again where this turns out to be valuable knowledge. Even a failed experiment if the problem is interesting enough is fruitful for improving your thinking about hard problems. Once you get past being disappointed over the experience, you'll probably find that a second attenpt at cracking the problem gets you further.
Once you have a currency that isn't controlled by any one single government, then they all have to behave themselves better with respect to financial management. That is the real point of one of these schemes - to make all the governments more accountable through the global economy. If you have a gold standard, then when a country is badly managed, much of the privately-held gold naturally flows out of that country, punishing the misbehaving economy. Likewise well-managed economies tend to have large inflows of gold. Governments hate this because they're in the business of having the final say in their respective geographic regions of control.
It should also be noted that in a real crisis, getting gold out of a gold hoard will prove problematical, especially if the vault is geographically distant from you. Likewise storing a couple tons of gold under your bed isn't terribly practical. Good luck trying to transport that if you need to leave in a hurry.