Thankfully I am working on the frontend side of things, but this is going to be hell when when we launch.
Thankfully I am working on the frontend side of things, but this is going to be hell when when we launch.
On the other hand, there's another project, which is a monolith. Around ~15 engineers. Incidents happen very rarely. The infrastructure is simple: just one beefy server without k8s (but there's a standby replica of course), so the team often does maintenance of the server on their own instead of a separate SRE/production team. The monolith is modular so it enjoys same advantages as microservices when it comes to module isolation etc. Highload functionality is moved to separate Go services only when needed (currently 2 or 3 satellite microservices). All changes happen in the same codebase so features are released faster and it's easier to reason about.
I'm currently on the second project and whenever I attend architecture reviews of the first project, I'm shocked how overcomplicated the first project has become. A simple addition of a new object property can end up in 5-10 engineers arguing for hours whose microservices are affected and how they should best communicate.
Engineers need something to talk about in their OKRs. One strategy is to refactor things into microservices.
Honestly, given the opinion of frontend on this forum, I'm surprised that you're thankful! I am 100% certain everyone here would take any kind of backend work over the hype-driven mess that is frontend development.
Comparatively, there are far fewer backend positions available that are more than basic CRUD, even if it those CRUD operations are gussied up with a bunch of needless complexity juggling sources of truth across a bunch of different services.
There's plenty of drudgery out there, and some interesting work in every field, but on the whole I personally have found frontend product development- product configurators for engineering / manufacturing, network visualization, visual design tool development, and data driven interfaces to be more fulfilling work.
To each their own, but maybe you might want to be a little more aware of the people that you work with when you feel like saying something condescending.
Edit: FWIW, you'd have to be blind to not see hype driven development in backend, especially in HN. "Rewrite it in rust" is the meme that continues to keep on giving.
A dozen micro services per developer on rickety infrastructure serving data for a pixel generator built on top of 5 years worth of JS hype cycles. Why are we doing this to ourselves? It is just so stupid.
The back-end is becoming an even more hype-driven mess these days. What with k8s and dozens of micro-services and controllers and telemetry-everything. And people don't even listen nowadays if you try to propose something simpler.
The front-end is at-least constrained by browser limitations and the client device. One can only do hanky-panky upto to a point. But at the backend, "throw another cloud machine" at it has become the de-facto solution for most performance issues.