In 2007, before YCombinator and AngelList had changed the industry, I worked in a nondescript office park in the heart of the venture capital industry off Sand Hill Road. Amid the leafy sprawl of buildings next to 280 and Stanford University, billions of dollars were and are invested out of the fancy offices of VC/PE firms you’ve never heard of. The whole industry has been shrouded in opaqueness since it was created decades ago, built on relationships from business schools, professional networks, and investor referrals.
This is an interesting response to Elizabeth Yin's post on why you can't find a technical co-founder that she wrote on Andrew Chen's blog.
She claimed your best chance was to learn to code and build your MVP on your own: this will not only ensure your MVP will be an MVP but also that you learn from it and in the process be more credible to and respected by your future CTO. Thomas Kjemperud doesn't disagree with the second part but claims there's no need to learn to code to do so: with all the open API's and existing tools out there, you should be able to build something semi-automated without even having to learn to code.
He's got a good point: to get 50 users to really love your service or product, do you really need a completely automated piece of software? Or do you need tools that let you deliver the value prop and makes sure it makes those 50 people happy?
It probably depends but the important thing is that you have to learn from these 50 before you do anything else anyway.