- Rosishi's Startup Diary
- Posts
- Change of plans, I'm quitting my startup idea
Change of plans, I'm quitting my startup idea
My startup is going on hold for some time, but I'll keep building still.
Change of plans…
As some of you already know, I started this newsletter to document my journey while building a startup “Dummy“ - all in one software for Web Designer (that’s all I can say for now). But, at the start of this week, after finishing UX & UI design for some main features, I decided to put it on hold.
Why? After dissolving all features in smaller components, I realized how complex it actually is, and that I’ve bit more than I can chew. I am a full time designer, slowly acquiring coding skills, which means that this would be challenging for me now. Don’t get me wrong, I love challenges, but, with an app as complex and useful as this, I want the build quality to be the best it can be, which I can’t 100% guarantee right now.
That being said, what’s going on now?
As I said, I believe I need to take my coding to the next level, and what better way to do it than to code. So I started a little challenge for myself:
Code one simple tool every day. That’s it. No specified time. For a year hopefully.
Point is to pick one specific problem, and create a tool that solves it (already existing tool or not). Easy to code, but still challenging projects. From simple Image converters to more complex Chat bots.
Every Saturday in this newsletter I’m going to update you with what I created. Today is day 4, and here are my tools so far:
Day 1 - Background Remover
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/cfdd00cb-3a73-4003-8f89-2e6424fdb470/backround-remover-screen.jpg?t=1720859703)
Wow. This was a fun project. For tech guys - all my projects are in HTML, CSS, JS, PHP & MySQL, additionally, this project used remove.bg API. I am still learning my way around API and JSON, so it was a bit though trying to work it out and find right documentation, but after that, everything else went smoothly. I just need to add another addition, as right now, every upload, and generated photos are getting saved in the app itself - I have to create timed function to keep deleting files from the server.
Day 2 - JPG Converter
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/53896c23-9595-4df2-9a06-cd3a4150190c/jpgconverter-screen.jpg?t=1720859979)
As you can probably already tell, I tried keeping same styling through out all the tools. I will aim to keep one type of layouts/colors for different categories (Image related tools, Text related tools, etc.)
This project wasn’t that hard at all actually, but it’s one I’ve personally been using sometimes. PNG to JPG, and vice versa don’t even need this, as you can just rename it with different extensions, but that ‘trick‘ doesn’t usually work on all the other ones - so, this solves it perfectly.
Basically, all it does is uploads the photo, gets content from it and packs it up in new file with wanted file type.
Day 3 - AI Image Generator
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/bcf7c86a-ba80-4bd1-a229-0d73383da75c/image-generator-screen.jpg?t=1720860269)
Even though it looks like the simplest project, it was honestly the most challenging one for me, as I had to use DALL-E 2 API from OpenAI. I have had experience with simple, text completion GPT API, but this is on another level. I am still not completely happy with the results it gives, as I have to specify a prompt even more probably, and make sure it always gives results as accurate as it can.
Day 4 - Image Caption Generator
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/51211d38-0d6d-4e9d-aade-b2ae295493ce/captiongenerator-screen.jpg?t=1720860471)
This is the project I did today, and I can say, it’s the best one so far. Both the way it works, and the fun I had coding it. Tool is using OpenAI Vision API, and generates 3 different captions, with two of them being oriented towards Social Media post captions, and even using accurate hashtags. I am actually surprised with how accurate it’s working.
So far, I really enjoy this challenge and I’m looking forward to creating more and more useful tools, and eventually publishing all of them. If you want to see daily updates about this process, you can follow me on X(Twitter).
Thank you so much for time you took to read this. This is first time ever that I’ve written something like this, and I am open for any feedback.
Best regards,
Nikola