At the end of last year, I was thinking about an ambitious goal: to create 12 projects in 12 months. I know that many people have done this challenge, and I thought it would be a good idea to improve my coding skills and find some new sources of income. I always have many ideas to build, so why not to try to do it?
A few weeks before these thoughts, I successfully launched News Poetry—a tiny web app that grabs news from The New York Times and creates poems from the headlines. This project received some attention, and I thought it would be cool to continue working with AI and poetry. That’s how the idea for my first project appeared.
I thought it would be cool to create a web app where a person could log in, choose the occasion, and generate a poem, which they could send via email or whenever they wanted. I thought that there are a lot of people outside the AI bubble who would be ready to pay a couple of dollars for an individual poem for themselves and their family.
For the first few weeks, I spent time creating the app itself and tuning the model. I liked the result, so I started to work on the landing page. I make up an awesome name — Poetronika — and bought a domain, and after I created a simple design, added some effects to photos, and it was ready to go.
After it was ready — I thought I was on time since it was the end of January — I started to work with Stripe and Supabase. And here things started to go off track.
I had never worked with Stripe or Supabase before. And to be honest, I never had experience working with databases on my own projects where I had to organize everything from scratch.
So, I started to explore, using ChatGPT to help me. And that’s exactly how I spent 4 months, lol. First, I struggled with Supabase’s policies, then with authorization (Supabase has weird documentation, which is partly outdated, partly in vanilla JavaScript, partly in TypeScript, and overall extremely messy). Then, I had some difficulties with Stripe and its webhooks.
At one time I was so frustrated that I don’t ship anything, that I came up with the idea of Dive Deep Book in 24 hours, just to prove myself that I can do something.
In the end I could solve all the problems — some by myself, some with the help of my friends, but it took a lot of time and energy.
So after that, the time for a soft launch started. I mentioned Poetronika in my newsletter and on Twitter, added the link to different directories, and promoted it a bit on Reddit. After that, I decided to attract my ideal customers offline. With the help of my wife, I created and distributed leaflets with a poem and a QR code to local coffee shops. The cool thing is that people were actually scanning and visiting the website. But no purchases.
And in the end, when everything was ready, I understood that I didn’t like this project. It just didn’t fit my vision of what I wanted to do, and I didn’t believe in it anymore. And then Apple teased that in a few months, each MacBook would have a function to create a poem from any text.
So now I don’t work on it. I still think that the website has a nice design, and I really like the poems it makes, but I am not sure that there is a market fit or if I want to work on marketing this project.
BUT! I learned a lot. Like, really a lot. How to work with database, how to connect the web hooks, how to apply policies. And the most important — I should ship fast to make sure that I am still motivated and have my initial vision.
I do hope, that all this will help me with my next projects.
Leave a Reply