Show HN: AI Domain Genius (aidomaingenius.com)

5 points by hennekec ↗ HN
Hi HN -

Last month I attended MicroConf in Denver and met domain investor Andrew Allemann of Domain Name Wire. I know nothing really about that space, but of course the topic of AI came up. He mentioned that he’s tried a lot of domain name generators, and very few provided decent results, and the ones that did (using AI) didn’t show if the domains were available. I thought this problem could be solved pretty easily by combining the ChatGPT API and GoDaddy API into a single, easy to use app.

A few weekends later, AI Domain Genius was born: https://www.aidomaingenius.com

(And yes, the app named itself!)

I’d love to get people’s feedback, in general, and specifically:

1. Do you find it easy and intuitive to use?

2. Does it generate any domain names where you think, “Yeah, I could see buying that domain for my project / business / website”?

3 comments

[ 3.7 ms ] story [ 19.2 ms ] thread
On a related note, since I’m still waiting on my Copilot X beta invite, I decided I would try to write the code for AI Domain Genius, as much as possible, using ChatGPT itself. I wound up discovering a fairly decent process where I start by asking ChatGPT to write a function with very basic functionality, and then keep testing the results and iterating with ChatGPT to add more functionality until I have a full module that does what I want.

Here’s my ChatGPT conversation to write functionality to save people’s favorite domains in localStorage: https://storage.googleapis.com/gqueues-static/chatgpt-favori...

ChatGPT definitely needed my thinking to guide it, but I ended up with roughly 40% of the front-end Javascript code written in this way.

1. Yes, very. Loading indicator could be a little more clearer on time estimate, but great/outstanding overall.

2. No, it seems too direct and not creative. I wonder if the original non-instruct models (curie etc) would be better for this use case than ChatGPT. Because they make more divergent and creative results.

1. Cool, thanks! And great suggestion with the loading indicator. Makes sense, especially since it does take awhile for all the ChatGPT/GoDaddy processing to happen.

2. Thanks, that's good feedback to hear. I hadn't thought about trying the different models. Currently it's using gpt-3.5-turbo with the temperature set to 1.3. While testing, when I bumped the temperature up higher I started getting a lot more non-sensical words / letter combinations. I do have a lot of extra guidance added to the user's prompt before it gets submitted to the API (including to be creative), but it sounds like it's still not quite there. I'll do some more experimenting with prompt pre-processing, and try out some of the other models too!