I made a similar leap from an OK-paying regional job at an established niche cashcow to a London start-up. The perks: a coffee machine, senior pay cheque.
The reality: a shit coffee machine, low moral, no realistic prospect of a profit and no willingness to pivot. 80% of costs have been sunk on a Facebook clone 'with a spin' while the owners try to sell snake oil to investors. We operate like the bad slides of 'Good vs Bad Startup' are a blueprint for success. The owners, almost weekly, come up with an idea someone has to coach them away from - a long ironically drawn out confrontational meeting of: "we don't have the resources". The bit I expected to pay off (medical monitoring hardware) turn out to be crappy I2C/SPI bridges any e.eng graduate could whip out in a month :(
I've had long look at myself: how could I turn down other offers yet accept this?! I'm still unsure why. For a while I feared that I deserved this - I'm one of them, one of the guys who decided to create their own header based HTTP authentication system key by a timestamp: a crap programmer.
It's been 4 months. Poor tests are still committed despite my best efforts to teach the one doing it that they should test a result not implementation (only 1 test damn it!). We don't use any JS frameworks on the front-end (yet alone my other true love: Typescript) because the last lead dev couldn't understand the immeasurable benefits of model binding. I've done - by no means single handedly - an incremental rewrite of the entire code base (front, middle and back). Although they're paying me on time, they missed my post probation increase :( I could go on but this is probably not the right forum.
In short: it's not worked out for me either. Thankfully I've got an excellent track record so I'm off in 3 days to a proper PLC. I heard large companies, like the BBC have 15 (FIFTEEN) designers in their News division alone... I can't wait to have just one designer unencumbered, available and talented.
The whole saga has left me with the strong suspicion most startups are a joke (no I don't want to be a DBA/CSS/JS and C# guru who maintains the iOS codebase!).
Inept owners who're unwilling to pivot, often trapped by the sunk cost fallacy, or owners expect to create a market with their 'one true solution [to a problem you no-one has]' plague the startup scene. No product should ever be a Facebook clone :(
For what it's worth, if I was hiring now I'd interview you solely on your written language skills (which are better than most native English speakers, myself included).
So if the benefits can't be measured, you're suggesting that the front-end should be rewritten as a matter of pure faith in model binding?
I can't wait to have just one designer unencumbered, available and talented.
Good luck with that at a large corporation, I guess... :)
Recently, especially at a startup I've felt the need to use libraries and frameworks more and more just to hit deadlines and meet the desired user experience.
Model binding is a prime example of this: on the web today everything has to be dynamic and reactive. Rather than do something (i.e. animate) and tightly couple the onComplete handler to N other things to do afterwards (via callbacks/promises etc) I prefer to compose my code around n-event subscriptions and react to changes.
(P.S. one designer is a step up from none!)
What I find is that each framework works really well for about 80% of a project, and then the last 20% get really painful. I haven't found anything that's elegant, flexible, simple, and native.
So unless this startup can start where FB is right now, no one will "clone" FB.
So replacing FB means recruiting some niche of customers, and providing something fresh and useful. That can be done without unlimited perks.
With thinking like that you will never displace the #1 company, because you're not thinking ahead of them, you're only ever trying to catch up.
It's much more likely that they exploit some new thing.