Ask HN should be two words rather than one. Having said that, the best answer I can muster up here is "it depends". Without knowing what you're working on spec

Ask HN should be two words rather than one. Having said that, the best answer I can muster up here is "it depends". Without knowing what you're working on specifically, who you are, what your situation is, etc etc etc... The following advice is meant to be vague and general so take it with a grain of salt. You should always validate your idea with committed users prior to even writing the first line of code. You should validate whether or not there is a large enough pool of people wanting what you have to offer and the market size. This goes without saying that this step should be done regardless if this is suppose to be the next big thing or just a simple side project. Assuming you have this done and want to jump into building the MVP, usually, its unlikely that your MVP will involve heavy ML in it. Giving you ample time to build, improve, and scale before you need to really do real heavy lifting (usually, not always). That said, if you've never done it before, there are lots of resources on here and other places (a simple Google search will yield results for you on previous posts) of how best to approach this. However realize its not going to be a quick solution and you won't master it in a short period of time, especially since you're involved in a new startup and theres a lot to do and a steep learning curve on a lot of other things as well. As for time frame? Thats nearly impossible to answer without knowing what you already know, who you are, how fast you learn, etc... Realistically, the likely answer is not very short albeit you can learn some useful stuff in a short period. As for interviewing someone with ML chops when you don't know it yourself, that's difficult advice to give. The best recommendation here would be to find people you trust and get recommendations and go from there. Its extremely hard to evaluate anything without understanding it yourself. Ideally if you work hard on learning ML, even if you're not great at it in a short time frame, it will help ease the process for hiring purposes and evaluating to some degree.