[rdl]

If your project failed, and you can take some time off, go or it. I'd suggest joining an obviously-successful project for a year or two after (a "unicorn", although they're not all great, is a good choice; a successful team at a successful company is, too), then doing a startup again if you feel like it.

Taking more than a few months off might be counterproductive (and expensive), and getting trapped doing consulting/contracting doesn't contribute as much to doing another startup again later.