Creating a product from a hackathon
After publishing https://quinto.cat in January (explained in this post), I was thinking about what I could develop on Februrary.
As mentioned in the last post. There was a hackathon sponsored by Cohere, an AI tool, which I decided that I would participate and use that project as my second project of the year.
The hackathon was due on February 7th, so by that time I already had an MVP of what could be the final product.
What is the new product and how is built?
Was a website built in PHP, HTML, CSS, JS and MySQL that allowed the users to search reasons for anything.
The users would look for reasons, and the website would return a list of 5 reasons for the requested topic. You can check it in https://www.extrareasons.com
Was quite simple, but at the same time challenging to find the best ways to generate content minimizing the use of the AI API to avoid any costs and be able to work with the free tier.
When I submitted the project, I didn’t know what else to do for the rest of the month, so I kept it like it was for the next week and a half.
After some time without doing anything about the project, I thought that I could maybe generate some posts using another AI to have some pages and try for Google to index them.
If I had more pages, I had more options for some users to come.
Once decided, I scrapped the most searched queries on google about extra reasons. And I came up with a list of more than 400 keywords.
With that list, I prepared a script that generated a post for each of them and saved it into the database, so after a couple of hours (the script had timeouts to avoid being banned from the AI tool) I had more than 400 posts on the site giving reasons for lots of different topics.
Also, I focused on the meta tags to be more “SEO friendly” and created a sitemap with all the posts for google to index it easier.
Once everything was done, I had a “good” product, but no users and no ways to get them.
So I opted to create a twitter ad campaign to try to reach some initial users. I didn’t want to spend too much, at the end I spent 10€ in total for the ads.
With that, I had a site with lots of urls, curated content generated by an AI, and some initial users…
What’s next?
Now, I plan to let it grow “alone” and let’s see what happens in a month from now… If there are some users accessing or if it dies on the internet sea.
Until then, I’ll work on the next project. Actually I already started this Friday on the afternoon, and I hope to have something soon.
It’s not the next big thing, but maybe I can get some users for it.
If you read until here, thanks for your time and do not hesitate to contact me on twitter @txordy to discuss about anything.