DEV Community: labdays

Vibe Coding Challenge — Day 16: Printable Designs

labdays — Sat, 14 Mar 2026 23:49:51 +0000

Bullet journals, to-do lists, habit trackers. All just a click away.

Announcement
With the application I'm releasing today, you can create productivity visuals using planners, checklists, habit trackers, Notion-style templates, and many more options, and then print, use, or sell them. If you'd like to try it, the link is below 👇

https://printcraft.labdays.io/

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 16th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 16th day of the Challenge
I've accumulated too many complex notes during the development process. These are mostly new ideas. My first task this morning was to clean up these 3000 lines of notes. It's strange to realize that things that seemed like good ideas when writing the notes are actually nonsensical when read later. Written during a dopamine rush…

In the AI era, I think our role is proper orchestration. There are things that do everything, but to create a meaningful product, you need to connect them correctly and in the right order.

Building in public has a certain power. People aren't very good at keeping promises to themselves.

AI agents are a powerful army for people who struggle with execution but excel in ideation and thinking. I've found another intrinsic goal for my project: to show people what can be done in a single day.

Vibe Coding Challenge — Day 15: Settings Generator UI

labdays — Fri, 13 Mar 2026 22:59:16 +0000

Claude Code, Cursor, Next Config — all the settings you can think of on a single page.

Announcement
The project I'm releasing today arose from a personal need. I got tired of searching for config file snippets for the Vibe coding tools I use and the technologies that need fine-tuning. This is how my project called "Settings" came about (yes, I think the name is very creative and original). It's completely free and contains plenty of config interfaces. If you'd like to try it, the link is below 👇

http://settings.labdays.io/

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 15th day. You can visit my website labdays.io to learn about the process.

Notes from the 15th day of the Challenge
Andrej Karpaty's microGPT project is inspiring. Actually, massive projects can be micro-sized if desired, just like AI models are distilled and made as small as possible. In short, distillation is all you need.

Claude Code continues to struggle with large files. Every time, it reads a 200-line section and summarizes it for itself. While breaking files into smaller parts increases complexity for humans, it's more beneficial for AI because it sees the big picture as a map and doesn't experience the lost-in-the-middle syndrome.

Claude Code constantly asks for permissions, which annoyed me, so I looked for a setting that would give Claude Code all the permissions. Then the idea for this project came to me, and I brought it to life. The beauty of the interface and structure made me want to improve this site a little more, and I built a settings generator tool with a UI using the current documentation of nearly 50 popular tools.

It has become one of the few projects I've done so far that I am proud of and believe will be useful. Now it's time to go to bed. It's been 15 days, wow!

Vibe Coding Challenge — Day 14: Landing Page Generator

labdays — Thu, 12 Mar 2026 18:20:36 +0000

Would you like to see what the project should look like before coding it?

Announcement
Today I'm releasing a landing page generator tool that allows you to generate an image of how an idea looks before bringing it to life. It's a tool that generates an image in the dimensions and length of a website, incorporating project details you might not have even considered. You might want to try it, especially in the vibe coding process, to see if it's a sensible idea before starting a project. If you'd like to try it, the link is below 👇

http://landingpage.labdays.io/

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 14th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 14th day of the Challenge
Claude is getting smarter on the web. Releasing a new project every day is becoming increasingly enjoyable. If I wanted to, I could spend months on those projects and keep polishing them forever. But I won't.

I'm trying to develop agentic tools like Openclaw. It's like there's a ghost in my computer…

The project I'm working on in the background has excited me a lot. The projects I need to publish now feel like they're what my boss expects of me. But the motivation for this process comes from this challenge. The human brain's thoughts and desires are shaped by whatever you fill it with. I care about artificial intelligence and the possibilities it can offer now. I listen to AI podcasts all day on the go. Every day I examine product hunt projects and improve my understanding by reading tutorials about AI models. Just like a Vibe coder should do.

As the codebase grows, working with AI becomes more difficult. For this reason, it's more logical to leave the interface for last and develop the front-end separately after building a backend application that runs in the background. AI has boosted my confidence, and I'm also working on a web3 project, so here's a shitcoin for everyone from me.

Vibe Coding is the most motivating tool for improving reading and writing skills. You just say the word, and it's done.

Developers keep creating to impress other developers. Most product hunt projects don't reach the end user, the real people who go out, earn money, and shop.

The faster a tool works, the more magical it seems.

In Huggingface, examining what very unusual models can do develops the imagination about what can be done.

Now, no force can make me write or read a single line of code. English is truly the new programming language.

AI is very good at starting projects, but not so good at finishing them.

I wasted too much time today and almost broke the series. I need to rest a bit.

Vibe Coding Challenge — Day 13: 1000 Different Perspectives

labdays — Wed, 11 Mar 2026 07:43:58 +0000

Ask a single question and get 1000 different one-sentence answers from some of history’s most important figures.

Announcement
The Perspectives project I developed today was something I've had in mind for a long time and something I personally needed. In topics requiring different perspectives, artificial intelligence usually offers only one perspective. But what if we assigned it 1000 different personas and had it generate each response individually? I think it turned out perfectly. If you'd like to try it, the link is below 👇

perspectives.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 13th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 13th day of the Challenge
I dedicate more than half of my day to a large background project, so the daily project I share is just a smaller, supporting one that I usually complete in 2–3 hours.

Ideas evolve; once they reach a certain point, they give rise to new ideas that only then become apparent.

Comments about 'AI slop' annoy me. Why do the exact same people who preach doing 10 push-ups or reading a single page a day for consistency hate the fact that I build one project every day? Yes, I think these are small projects, but they will get better and better.

Bodybuilders who use steroids and vibe coders are similar in a way. Yes, steroids speed things up, but there's still a lot of hard work involved!

I've found a way to constantly improve the project I am developing in the background. It's not actually a secret, but most people don't do it. Test, fix problems, iterate. Once all tests pass, make the tests harder. The feedback loop is very powerful.
If something isn't good, increase the quantity. This will make it look better. Big is different.

I learned about a concept called Combinatorial Creativity today. What AI does today isn't strictly about generating original creativity from scratch; rather, it is based on combining multiple elements to create something new. Thinking about it this way, human creativity isn't entirely about creating something out of nothing either…

I'm thinking of doing RAG and fine-tuning on AI models.

Vibe Coding Challenge — Day 12: Tattoo Generator

labdays — Tue, 10 Mar 2026 05:53:11 +0000

Create custom tattoos in dozens of different styles.

Announcement
Today’s project is a tool that can generate beautiful examples in all tattoo styles and help you find the right design by allowing you to describe what you want before getting a tattoo. Choose your style, describe your tattoo, and it will create a printable tattoo design for you in seconds. If you’d like to try it, the link is below 👇

inkmachine.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 12th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 12th day of the Challenge
When Vibecoding, it’s easy to spot an error in the interface, but it’s difficult to spot an error in the backend.

Now, it’s not enough to just have a great idea and turn it into a product. Even the most complex and monopolized software can be imitated in a short time. Considering that a framework like NextJS was rebuilt much faster — in just 1 week (by Cloudflare) — things are getting increasingly difficult.

Now, there are only minutes between ideas and implementation, not days, weeks, or months as before.

Architecture and orchestration are becoming increasingly important. Using powerful tools with the right algorithms allows you to turn their power into a usable product.

It’s beneficial for someone doing vibe coding to have knowledge of topics like machine learning, as it helps you better anticipate how AI models think and how they should be prompted.

Vibe Coding Challenge — Day 11: Road Map Generator

labdays — Mon, 09 Mar 2026 06:48:25 +0000

Discover the path to your dream.

Announcement
Create a roadmap to becoming the person you want to be with the roadmap generator I released today. Moreover, generate the path to your goal with a single click by providing resource links and in-depth sub-branches. If you’d like to try it, the link is below 👇

roadmap.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 11th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 11th day of the Challenge
Unfinished tasks from the previous day take a toll on tomorrow’s productivity.

Another answer for those who ask why I do this: most projects are released as beta, and I put extra effort into growing the projects that attract users’ interest. The projects I release are to continue the series. The projects I develop in the background are for real products and are more carefully crafted.

AI is not a bubble, but no one knows exactly how to use it most efficiently. The most valuable output it can produce right now is code. It will be much more useful when it becomes embodied in the future.

Context and memory problems are still among the biggest problems of artificial intelligence. Instead of expecting it to retain a huge text, it is possible to compress a large context into subcategories.

No effort in life is ever wasted. Even if there are no direct rewards, there are bonus consolation prizes.

Synthesizing unrelated topics and trying to transform unrelated things into each other is useful for finding creative ideas.

One of the biggest problems with AI today is cost. If it were cheaper, we could use it much more freely and become much more productive.

I understand Edison better now. It’s hard to find 10,000 ways that won’t work.

Vibe Coding Challenge — Day 10: Know How Generator

labdays — Sun, 08 Mar 2026 06:01:20 +0000

Announcement
The project I'm releasing today is a tool for creating well-formatted, visually-supported rich content in today's world where content creation can be automated. Instead of searching for "how-to" guides on any topic, discover how to do it directly. Whatever you want to ask, it provides you with a step-by-step guide. If you'd like to try it, the link is below 👇
knowhow.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 10th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 10th day of the Challenge
My projects receive various reactions. Some support them and tell me to continue this productivity. Others hate everything that is AI-generated. Doing a lot of experimentation contributes to the creation of increasingly better products.

The mini/beta projects I share lead to and evolve alongside larger projects I'm developing in the background.

It's essentially an experiment. Currently, my projects are at a lower-to-medium difficulty level. However, I will increase this level over time, and then people will better understand what can be done with AI in a single day. And we need to increase the scale of what can be done in a day and push our boundaries. This has accelerated my pace in my personal life. It can inspire other people to accelerate as well.

Every day, I generate 50 ideas in my notebook. Some are very absurd, some are very difficult. But no one can keep generating bad ideas for long. Quantity begets quality. Waiting for inspiration to achieve quality leads to analysis paralysis.

Vibe Coding Challenge - Day 9: Novellasense Book Writing Assistant

labdays — Sat, 07 Mar 2026 11:58:04 +0000

It's time to finish your unfinished novels.

Announcement
The book writing assistant tool, Novellasense, which I released today, has a minimalist structure that will allow you to finish your books quickly, whether you want to extend sentences and paragraphs or rewrite a section differently. You can auto-complete with the tab key, just like writing code. You can also set the style and direction of your book in the AI Directions section.
If you'd like to try it, the link is below 👇
novellasense.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 9th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 9th day of the Challenge
Every revolution devours its children first. Artificial intelligence is devouring programmers.

Actually, I'm conducting numerous experiments. Even though I'm only publishing one useful product, my pool of ideas is constantly growing.

Coding AI and prompt-writing AI should be on separate platforms. When I tell an Agentic AI what needs to be developed in the project, it takes shortcuts. However, if I plan with another AI, it works much more efficiently.

My parallel projects always influence and reinforce each other, so I try to start more than 5 projects a day, and a good product always comes out of it.

Today I went out of town and could only work half a day. Even the slightest interruption slows my momentum. So I mustn't stop.

If you let it, a project will last forever. It will bring you to your knees and keep you there forever. Neither you, nor I, nor anyone else can outlast a project.

Vibe Coding Challenge - Day 8: Case Files Dedective Game

labdays — Fri, 06 Mar 2026 06:04:19 +0000

Each one is a unique and different detective game. Artificial Conan Doyle

Announcement
My detective game, Case Files, which I released today, seems to be a good example of this type of game in terms of visuals and interface. You can create case files on any topic and content you want, collect clues, and examine suspects like a detective.
If you'd like to try it, the link is below 👇
casefiles.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 8th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 8th day of the Challenge
AI is quite good at the frontend, but when it comes to the backend, it gravitates towards stacks with external dependencies and requires dashboard setups like Supabase.
I'm using Vercel for deployments, and the backend operations are resource-intensive. I need to find a more cost-effective stack and deployment option.
The best thing about Vibe Coding is that you don't even know the features on your site. You just think, "Wow, I hadn't thought of that, it's amazing!"
I admire the work that comes out. However, I'm also curious how far it can continue to be developed. It leaps 100 steps forward initially. However, when you try to build upon it, it cannot linearly add another 100 steps. Its capabilities decrease as the codebase grows.
Doing a project every day is really tiring, but also satisfying. I couldn't continue without the dopamine it gives me.
People are slowly starting to notice my projects. I need to produce more advanced work and develop open-source projects that everyone can use.
I'm realizing that everything digital is starting to become worthless with artificial intelligence. It is not only daily projects like mine that are losing value. I doubt even the projects that are number 1 on Product Hunt are making any profit. I think nowadays, developers are creating for each other in the digital environment. Ordinary people are completely oblivious to all this.
Digital products have become devalued. We need to try to create things that bridge the gap between the digital world and the real world.
The more connected something is to the physical world, the more valuable it becomes.

Vibe Coding Challenge - Day 7: Web to Book

labdays — Thu, 05 Mar 2026 06:17:48 +0000

I have always wanted to be able to read websites like a book.
Announcement
With my WebToBook project that I released today, you will be able to convert any website you want into PDF or E-Book format with a beautiful book-style layout.
Whether it is a single-page link or multiple pages made up of several links, all of it will be formatted with proper book typesetting and presented with a clean, well-designed layout.
Additionally, with the auto-include module, you can choose whether the system should automatically detect and include subpages or related pages of the page you provide and add them to the book.
If you'd like to try it, the link is below 👇
webtobook.labdays.io
Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 7th day. You can visit my website (labdays.io) to learn about the process.
Notes from the 7th day of the Challenge
I wrote the article from yesterday and shared it on 10 different social media platforms. Meanwhile, I had two projects built in the background using their initial prompts. I'm learning more and more (multitasking and working quickly).
I don't bother thinking about project ideas; I turn even the wildest ideas that come to mind into projects to see what they will turn out to be. Instinctively, I want to sit and think for days about which project I should do. But now, I am a vibe coder.
Whenever I get stuck on project ideas, I hang out on Product Hunt for a while. I just get inspiration or discover different topics that can be worked on. I've set a goal to read 250 newly launched Product Hunt projects every day; this builds my idea muscle. Sometimes it reminds me of something, or I misunderstand what I'm reading and come up with another idea. Misreadings are sometimes very valuable.
Most sectors are achieving noticeably good results by integrating AI into their work. However, since AI is taking software development out of our hands, it's becoming the main thing, not an additional thing.
Everything can be improved one step further. Ideas that seem simple at the beginning can become much bigger than their initial state through development cycles. Big things have small beginnings (Prometheus Crew - David 8).
Everything is meaningful in its own time. If I could have done these projects 5 years ago, they would have been great achievements. Now they are ordinary… Being fast allows you to keep up with the times.
Some days Claude works very well. Today is one of them.
By noon, I had prototyped 6 different project ideas. Some of them are useful.
I've become bolder in prototyping new projects and their potential. I can bring to life project ideas that I've had in mind for a long time but never worked on, with just a few sentences, and get a working prototype. After that, it's polishing.
Instead of sending the entire prompt directly, putting it in a TASK.md file and telling the AI to execute the tasks there is very effective.
I have some great projects that specific audiences would want to buy, but I need to develop a membership system and use it as a template across my projects.
Things that used to feel magical are now commonplace.
Developing a project iteratively can take it very far from its starting point. That's probably why developers are called developers.
Big is different. Even if each of these projects turns out to be useless individually, as a whole, they will make a difference.
The agent concept actually relies on words. Things triggered by words that lead to actions…

Vibe Coding Challenge — Day 6: AI Calculators

labdays — Wed, 04 Mar 2026 06:31:42 +0000

GPTools — The point where quality and quantity meet.

Announcement
The project I launched today is actually a side product of my prompt engineering project. I have a structure that can generate high-quality prompts, and with this structure, I can create unlimited calculators and customized AI tools powered by tailored prompts.
It is an experimental project, and if you'd like to try it, the link is below 👇
gptools.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 6th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 6th day of the Challenge
This morning I finished the remaining work from yesterday's project, and now I'm getting started on today's.
I created an archive for the recurring prompts I use.
I'm noticing that as my projects progress, new, interconnected project ideas emerge (a kind of compound effect).
I started a project to develop on Openclaw, but I canceled it because it was still problematic even though it took me 3 hours to set up.
I've noticed that I'm gravitating towards simple projects because of my self-imposed rule to finish a project every day. I need to do more than just an AI Wrapper.
The bigger the projects, the more I want to do them. I need to stop dreaming small.
Imagination is a matter of courage.
Self-generating structures have become more feasible with AI agents. All we need are more tokens!
AI is about adding intelligence to things. Everything can be smarter.
Today I developed an extraordinary project, but it's not ready to publish yet. So, to keep the chain going and continue the challenge, I need to make an AI wrapper.
"Big is different." I love that saying. If something doesn't work, make a bigger version of it.
The actual purpose of this vibe coding challenge is to test whether AI is replacing my profession as a software developer. If I can't earn money or find work despite working on a project every day for a long enough period, I should revisit my dream of becoming a carpenter…

Vibe Coding Challenge Day 5: Custom Font Generator

labdays — Tue, 03 Mar 2026 05:43:59 +0000

Announcement
With my font generator project that I released today, you can create any font you want with a style you define. You can freely use the unique fonts of a movie or game you are a fan of. The link is right below 👇
davincifont.labdays.io

Context
I started the Vibe Coding Challenge. I plan to release a new product every day, and today is my 5th day. You can visit my website (labdays.io) to learn about the process.

Notes from the 5th day of the Challenge
Starting a new project is my favorite thing. The later stages and polishing are a bit more tedious.
Generally, some work from the previous day's project spills over into the next day.
Lack of sleep visibly reduces productivity (Going to bed at 3 AM - Waking up at 7 AM).
I used to spend a long time looking at the projects I had done and the articles I had written, thinking about them over and over again. Now I don't think about them anymore (I don't have time!).
This challenge taught me the importance and satisfaction of working fast in just 5 days. Even if I work on a different project in the future, I should replicate this fast-paced challenge format there as well.
The fact that AI can code saddened me as a developer. I felt like the workers who broke the machines during the industrial revolution. But now I love it and I am integrating it into my workflow.
I am noticing how my previous projects are compounding into my future ones, and this is amazing.
When I look at the old projects of the creator of Openclaw, we can see the small pieces that lead to Openclaw. I hope this will happen for me too and lead to a big project.
Yesterday I only slept for 4 hours, and this directly impacted my productivity. I took a 30-minute break, and even that was enough to recharge my energy.
Half the day is over, and I have 5 half-finished projects.
The initial state of a project is very important in terms of its progress. When overly detailed plans are provided, the project spirals into an unmanageable state through its own decisions.
It was disappointing to realize that even the projects I had planned and considered niche had already been built.
I suppose once everything has been built, what remains will be novel inventions and the synthesis of multiple things, transforming them through proper orchestration.
Some days the AI works well, other days it works very badly. Even if they don't reflect it to the user, I think they are secretly throttling the performance.
The more projects, the better. Some don't lead anywhere good, and it's better to have your hands full than to end up empty-handed. Days are short.
Although design shapes thought, functionality comes first. I shouldn't make the mistake of jumping into design without nailing down the project's functionality ever again.
Instead of providing a long prompt, it works much better when I point to the task.md file and ask it to perform the tasks there.
The project name is also an interface and changes the way you think about the project.
It's 6 hours until midnight and I still haven't produced anything tangible (Lack of sleep is not good at all).
Time constraints make implementing features like membership systems difficult. One of my next projects might be a membership template.