This is completely my opinion and shaped from my own experience. But if you're working at an understaffed startup, the bottleneck for 'success' is usually not the engineering team, and also almost definitely not correlated with butt-in-seat time and amount of code contributed. Not saying that's what you were implying, but just want to clarify that's the direction I'm coming from.
It's up to the business's leaders to determine what should be prioritized with the resources they have and take risks on building out those ideas accordingly. If you, as an engineer, begin working nights and weekends to have the company 'succeed', you're now communicating to management there's all this untapped engineering capacity. You give the illusion that more can consistently be done than is actually sustainable and no one wins when the engineering team burns out 6-12 months later. Well, maybe the company does somewhat when you leave and forfeit all your stock options back to the pool...
In my opinion, the biggest impact you can make as an engineer at any level is by stepping away at the end of the day, doing your own thing, and every now and then just think holistically about what you're doing at work and the direction the product you're working on is going in. Your butt-in-seat time will become far more productive as a result.
