The same as any small developer. Most of the time it'll be a part time deal... but there's no rule - part time might end up be the majority of their time, it might just be weekends. Either way they'll likely be something else helping it to exist if they've been going more than a year.
It's a lot more feasible these days with just communication/documentation/middleware and even online communities being far more pro-active than it they ever used to be.
As far as I know Grev was indeed basically founded by 1 main guy with a partner, but they've hired in and out services from day one - making full use of their previous Taito connections and relationships with Treasure and Sega in particular... I don't quite have this straight time wise in my head (let alone the sources) but I believe g-rev funded itself helping out on Ikaruga and then Gradius V just to support their own games (i. border down and ii. under defeat).
It's a little simplistic to say 2 guys made Under Defeat; even if they did most of the work - more people made it possible. Same I think with ikaruga - basically 2 guys and then a music guy made most of the game (I *think*-someone correct me if wrong) - but it was made possible by more. ... It's a bit like saying minecraft was made by one guy - the full picture is a little more than that
...
...
There's relatively few shump centric full time developers, or even hobbyists: Staff will likely be more like freelancers who own a company that holds their self made games and backs up their other ventures. Most often these small band types will just want to make good games rather than necessarily games of just a single genre.
You might find an individual or duo who dedicates themselves to a task for a 6 months/year, or so, since that's where they are in life - after which they'll likely finish off something in their spare time after doing money work - or if they manage to make enough cash with a title release/early access/kickstarter/publisher that they can start doing it full time.
A lot of younger developers, even with a decent demo, will land themselves a job/connection that will help support themselves. Inevitably this cuts into their time; but slow developments are better than none. Some lucky developers will find supporting funding/relationships or work that relates to directly to their project in some manner, but some will just be working on something completely different.
Often key creatives start to become more like producers/directors if they pursue the business management/development side of their own company. This maybe exactly the reason some developers remain small and working on shmups - to do otherwise; to proceed in another manner, may simply dismantled the key parts / motivation for the developer to exist and work on shmups to begin with.
Sounds silly? But consider this, you like shmups and enjoyed making one on the side with a bit of help from a mate/contractor here and there: you did ok and now have bit of cash. Now presume you want to make another shmup, would you take a year or two off and make a shmup yourself (perhaps quitting your current job/family setup)? Spend that cash on staff/loan to help make a company to make another shmup? Or a different kind of game/company to make more money and eventually fund another shmup years from now... maybe... ? Or would you be better putting that cash into a mortgage, and just making another shmup when you felt like it?... It's going to be a rare thing for all those paths to align to the point that someone maintains a shmup developer on a full-time basis
