We're quite far from implementing such a system for software "engineers".
We don't trust building ethics, independent inspector comes and checks if everything is as it should be before it can be used by the public.
Clients not being experts at the job they are getting somebody else to do is not a new pattern. So while some trust is required, it's best if you can get somebody else to verify.
I've seen a few smart clients over the years which when faced with some excuses from a software house hired another one to give them opinion about the codebase and capabilities. It seems pretty intuitive. It seems like a money well spent.
In fact, a whole lot of the time, the entire reason someone hires a professional is precisely because they themselves aren't experts.
Good narrative: but certification and guilds do not solve the problem.
For example: UK Grenfall towers. https://www.bbc.com/news/uk-61724373
It is a wrong to assume that only engineers can cause deadly mistakes. Also we have penal liabilities that don’t need licensing:
The [UK] Health and Safety at Work Act 1974 is designed to stop employers putting the public at risk, not just employees in the workplace. Individuals can be prosecuted and a serious breach could attract a two-year prison sentence.
And sometimes some pretty big exceptions: the [UK] government can't be prosecuted for corporate manslaughter
Locally to me in Christchurch, New Zealand, there have been no prosecution for the CTV tower collapse: https://www.nzherald.co.nz/nz/fatal-ctv-building-collapse-po... https://www.nzherald.co.nz/business/govt-considering-introdu... The second link is interesting because it looks at changing the law to add liability (not engineers licensing changes )In both cases, there are multiple layers of failure, and many causes could be asssigned. Especially the CTV building with inspections before collapse and warnings from people working there ignored.
You are talking Nirvana fallacy https://en.m.wikipedia.org/wiki/Nirvana_fallacy
If we assume there is no corruption involved, then lack of competence from the project management side can fail such a project. For this example it could be failing to mention or think about the extra load on the first hours in the SOA
Usually there is some sort of RFI process, where they ask a few companies 'hey can you build this for us? What are the types of services you would propose'. The list of companies here is already more or less pre-existing partnerships, or ex colleagues or...
(it mostly always contains Microsoft, and your boss is ex Accenture, so it involves Accenture, and for good measure to seem like they are open to other options they invite Deloitte and some other players as well, sometimes even IBM has joined the club again)
Then they decide who they deem thrustworthy, and you end up with Microsoft and (insert boss previous employer). So not only do you not get the lowest bidder, you can some veeerrryyy generic company that doesn't care and just sends juniors to solve it. This process is called the RFP. And it typically is far from neutral
The whole process is long and drawn out with all sorts of checks and balances built in to it, based on previous learnings from previous contracts that have failed in various ways (particularly if it has embarrassed an elected figure). No doubt on the back of this failure, there will be more conditions added to the bidding process, making it even harder to find a vendor.
Usually by the time the entire process is done, there's not many vendors left and in my experience they're usually not the ones you'd actually want to do the work if you had a choice, just often ones that'll at least get you something.
There are probably a dozen reasons why something like this might have occurred, and not giving the vendors a free pass, but assuming that a more expensive vendor would do a better job with security and reviews is just as likely to be a mistaken belief.
If a project is too expensive for a client to do well, they should not be doing that work in the first place.