3 ms·
Ask HN: How can I solve the chicken-and-egg problem of starting an AI startup?
I'm working on creating a statistical model to test the feasibility of my idea. But it's tough. I only have around 550$ a month to spend on training a model on the latest GPU's, which is barely worth a few days of training, and since I'm not an expert in the field (not even remotely, I come from a programming background), it takes a lot more trial and error to test things and make things work.
So I need an expert in the field and money to train a model on my data, in order to start up. But an expert in the field wouldn't join me without some credibility, which could mean having some breakthrough technology, or funds from an investment backing me up. But an investor would not invest in me without a team proficient in the technology and knowing the idea is feasible.
How can I go about solving this chicken-and-egg problem?
- sharemywin 9y agoCan you look at the potential customers your looking at solving a problem for and solve an easier problem for them.
- sharemywin 9y agoif you have exclusive access to the data then maybe you can partner with an existing firm.
- trickely_boom 9y agoI rather create my own company that is completely focused on this one thing, so I don't want to give another company access to my proprietary data.
- ilijalalkovski 9y agoI think that you should consult with an AI engineer to give you an insight that your idea is technically feasible. I suggest finding an AI researcher that will believe in your idea and join you as a partner, someone who can see the potential in your data. I have some experience in AI, maybe I can help you. Feel free to contact me.