The code there sucked - and most coders did, too.
I don't mean to give you the impression that your new job is going to suck. If you chose the right startup, you're going to have an awesome time. But I also think it's important to be realistic.
I think a big part of the reason a management tier is needed at all in a lot of smaller companies is the general unwillingness of engineers to sacrifice code purity for expediency.
Regardless, that's beside my point. It sounded like the person I responded to wanted to work on better code, so I was just pointing out that startups aren't necessarily any better in that respect. I'll leave it to him to determine how good or bad that is as I'd really rather not get involved in this old debate again.
I won't say the code was always pretty at the startup, but due to the higher level of skill among the developers, I think the code there was superior to what was produced at the big company.
Although the startup went under after I was there 2 years, I learned 10 times the stuff in 2 years of the startup than I did the previous 5 years at the big company. The skills I learned at the startup allowed me to get into a much better job at another big company that I stayed at for 11 years.
The point is this: build your software right. At the very least, only take reasonable shortcuts, and make sure you come back around to fix your crappy design at some point. The longer it festers, the worse it will get.
And it'll end up being 'production' code because "we already have it and they already paid for it!". And the crappier it is, the longer it will remain in production and the harder everyone will fight against rebuilding it to be not crappy.