8
10 Comments

My first launch took almost 3 years...

Hello there πŸ‘‹,

My name is Diego, and I am surprisingly resistant to launch.

I have a background in CS. I have been creating small projects for ages, but I have never made anything public...until now.

Reading indie hackers has made me realize that this problem is more widespread than I initially thought, but it has also motivated me to take the plunge.

I also suffer from some of the same illnesses many software engineers have: we love technology, playing with it, and finding better ways to build and optimize our projects, which often leads to a lot of wasted time and not much progress when it comes to launching.

Things that I never launched:

  • An app to automatically grade Python exercises
  • An app to generate funny pictures of friends using AI

The lowest-hanging fruit

After several projects ended with nothing, I decided I needed a different approach. At work, we always recommend starting with the lowest-hanging fruit: something straightforward that can become a quick win and motivate you to keep going. This contrasts with beginning with a super ambitious project that will become sidetracked and probably die when things get complicated.

I decided that it was time to take my advice, and I decided to build Will AI Do This? (willaidothis.com) It is a simple website to share prompt/answer pairs of interactions with LLMs.

I work as an AI consultant, and after sharing an experiment on social media, it was hard to find the example again.

The website is simple, took me about 10 hours spread over a week to build, and is now live!

Next steps

I don't think I can make money from the website per se, but there is a pseudo-plan for commercialization.

Since the setup is 100% free (except for the domain), I can leave it up and running forever, so that's nice πŸ™Œ.

If it somehow becomes a widely used website, I might try to find a way to do affiliate links or ads for AI products, but I'm not counting on that.

Plan B is to write a tutorial explaining how to build a website like that one, host it for free, and sell that tutorial on GumRoad or something similar. I might create another website using a similar tech stack to get more insights into what works and what does not.

Tech Stack:

  • DB: Google Sheets
  • Frontend: HTML + Tailwind + HTMX
  • Building Engine: Python + Jinja2
  • CI/CD: Github Actions
  • CDN: Cloudflare

Questions for the community:

  • Should I launch on PH, HW, or Reddit?
  • What is the best way to start building in public on X?
  • Is there anything else that I should be aware of?
on March 31, 2024
  1. 2

    You definitely should launch on these platforms and be open to feedback that you will receive. Also, have low expectations. Everyone likes a good surprise, anyways.

    1. 2

      I'll try, thanks. Low expectations is hard, but I guess is the only way to go

  2. 1

    Start fast and fail faster. The success of this project will be determined by what you want to achieve from it. PH, HW, or Redditt can help. I would start off with your friends, fellow devs, first. Grab some feedback.

    1. 1

      Yeah, I get that.

      It feels that, at this moment, I need a little bit more tracking and people using it. Maybe unbiased feedback could be helpful?

      How do I drive user adoption and make it more popular? That's what is mostly on my mind at this moment, beyond fixing bugs πŸ›

  3. 1

    I understand your situation perfectly! As a developer, I've encountered this scenario many times. It's common to want to wait until everything is perfect to launch, but the market often has its own timeline. Initially, I planned to launch the MVP of my product, (Modalcast), in just one month, but due to excitement and some setbacks, it ended up taking 4 months to complete.

    1. 1

      I can totally relate to that. How did you manage to keep working on it after month 2+?

      Did you manage to launch? Did you ever felt that you should learn a new technology that could be a better fit for the problem instead of using what you already know?

      1. 1

        What motivated me was that I was developing and learning at the same time, as the project made me face challenges I had never encountered before, always maintaining the same JavaScript stack, because with it I was able to accomplish everything.

        And yes, I finished the MVP! It's a widget that can be installed on any website. Check out www.modalcast.com

        1. 1

          Looks super interesting, congrats!