I have been through this (as the only tech founder in the group) three times with three different startups and found different reasons that it did not work out.
1. Equity. I like to think that the percentage of equity you get in relation to the amount of work you will be asked to do. 20% to do all the product (and/or service) development seems way too low for me.
2. In one of the startups I founded, we could not convince the non-tech CEO founder to release the MVP to early customers - even after extensive customer interviews, because the non-tech was not comfortable that it was an MVP. No desire to find the golden product/market fit and no desire to release anything less than a full blown, polished product (which we could immediately start charging for). And no funds for development either (just equity) - so entirely too much development needed to be done before any return (and no customer feedback).
3. And then I was in a startup that had a non-tech founder who was very weak at sales and very bad at raising money. But really good at asking for AI, VR, voice recognition, and (add buzz word here) in an MVP - entirely created and supported by a sole developer.
I would suggest staying away from any of the above.
I have found DO to have more additional features, but we use few of them.
The new DO K8 support looks interesting as we currently use Openshift (Red Hat), IBM, AWS and Google.