Those responses were from back in January, they're not responses to the current problems.
Apart from that, the 502, 503, 504 or what-have-you errors still indicate you are 'both' right and wrong about it.
It is a routing error, so it can't be the servers you're trying to reach themselves.
It is not happening for every user and not all the time, so it's not directly a Nexus issue.
It is, however, neither an issue on the users' end exactly, as it seems to be unpredictable for those as well. Sometimes it works, sometimes it doesn't, and for whom it works or not changes randomly.
It must be some router or proxy 'very' close to the Nexus servers, or it won't be happening to all kinds of people from all over the world.
If the Nexus does have a central gateway or something of their own, which distributes incoming requests to various servers, it could be 'this one' is overwhelmed by the amount and can't constantly serve the correct routes, or no routes at all. This is what those 502 and 503 errors mean, after all.
I know this from experience with what I'm working in.
No matter what I do, if I accidentally shoot down one of our servers, I will not be able to create a 502 or 503, only 404s can be produced that way.
But what regularly happened in the past was misconfiguration or overload on the one or two of the 'load balancers' placed in front of our servers, 'gateways' directing incoming requests through to one of our actual servers which is the least 'loaded' at the time (thus the name), causing such 502 and/or 503 errors as well. They mean, after all, a 'routing' error.
So, if the Nexus 'does' have such a gateway of sorts or similar between the net and the actual servers, then this would be my first place to look for the trouble.
If they 'don't', the blame lies on their service providers, out of their reach for intervention or fixes. This happened to us once in the past as well. One of the load balancers went off-line without our knowledge, and as such 50% of all requests went 502/3 without our doing. Turned out some maintenance technician of the serverfarm forgot re-plugging a network cable he removed for a maintenance task the day before. Needless to say they paid a 'reasonable' conventional fine to our big-business partner for the service down-time and loss in income over the time they caused by this.
I wonder why only 'features' of websites, that is Nexus-network-internal 'ajax' calls mostly, fail and not access to the sites in general so far. But looking at it from this side, there you could have the source of the trouble, the ajax calls or internal communication in general.
Nevertheless, I know the staff is already working on finding and fixing the issue with all they have, and they will have fixed it sooner or later. They always did in all the years I'm a member here. Sometimes it takes longer, sometimes not. But all it needs is patience. I wouldn't be a 'long-time' member, if they didn't always fix those issues in a timely manner. ...But the key here is 'time'. Give'm some, will you?