Malin Shirokanski (@MShirokanski) header
Malin Shirokanski (@MShirokanski) profile picture

Building Shard ($0 → $10k MRR) Documenting my vibe-coding journey. Sharing what I learn along the way.

Joined March 2026
I’m building Shard to $10k MRR in public. It’s an app for ADHD students who actually want to get sh*t done. The plan is simple, but the execution will be hard asf👇
5
9
779
X was supposedly going to start punishing copied content. It's been a month now since I saw that announcement. I open X, and I see only recycled posts one after another. It's gotten to a point where I can read the first couple of words and predict what is written in the entire tweet.
9
Has anyone else been losing motivation to post on X? The content that I've been seeing recently is repetitive and mostly just engagement farming. Growth is slow, and it feels like a waste of time when I could be working on my web app instead. I just don't see the benefit of posting unless your target audience is on X. Anyone else feel like this?
2
4
35
What are some skills that have become irrelevant with the emergence of AI tools? Some people would be quick to say coding, but I don't agree. I can't really think of any skills that have become fully irrelevant. What about you guys?
19
I used to not know what a git branch was. Now I add even the smallest change to a separate branch. The risk of AI breaking the app (especially if it's live) is too high. The inconvenience of one PR is nothing in comparison. Never push directly to main.
21
Has anyone here used AI UGC successfully, or is it all just hype? I'm seeing so many people promote this stuff, but I can clearly tell it's AI. I'm sure anyone under 40 years of age could also tell. I don't see how it can really work, considering people on social media aren't the biggest fans of AI-generated content.
1
50
Vibe coding gets such a bad rap these days, mostly because people completely misunderstand it. The haters always picture some lazy, unskilled person just sitting there typing, "Make me an app, and don't mess up." The reality is nothing like that. You’re constantly forced to make decisions on things you’ve never seen before, and you're digesting a flood of new concepts every single time you open your project. After a year or two of grinding, most serious vibe coders naturally pick up the core fundamentals of software engineering just through sheer exposure. And the grind doesn't stop after you launch. Once the app is done (or even before), you have to figure out how to market it, which is a challenge that a lot of the traditional devs who are hating on Vibe Coders don't have the courage to face. At the end of the day, the person who takes action and actually ships a product into the real world—even if it isn't flawless—is always going to beat the guy sitting in his comfort zone just working for someone else.
1
1
51
The harder you push to finish your app, the harder the app pushes back. Has anyone else experienced this issue? Plans never go smoothly; there are always 5 new things that pop up in the middle or after a plan is implemented. Maybe it's because I'm in the final stages, and that last 20% really is 80% of the work.
1
21
If you are posting UGC content to market your app, that specific account you post on gets "locked" to that niche. But I also want to build a personal brand as someone who builds and monetizes apps. So I decided I should make 2 accounts, one for my personal brand and the other to market the app. Is anyone else experiencing this issue as well? How is it going so far?
3
45
Is it true that some people can engineer virality, or is it a myth? What do you guys think? I believe you can never guarantee virality; at the end of the day, you are at the mercy of the algorithm. You can optimise for virality, but sometimes even when you do everything right, you might fail to go viral.
1
3
50
Why is Opus 5 so expensive, holy shit? I left it on for 10 seconds after the 5-hour limit was reached. I was at $1.72 used, and after the 10 seconds, it was already at $3.63. Crazy that Fable is even more expensive. Can't imagine how much money Fable would burn.
16
I asked this on Reddit, and it seems to be a consensus that nobody makes money from vibe coding unless they have technical knowledge. I have my doubts about that, so I'll ask the same here: Are there vibe coders in this community who can't code but make money from apps made with AI tools?
2
2
41
Builders, what do you prefer? 1. Working in the morning right after you wake up 2. Working in the evening/night time.
7
Hot take: Launching to 0 users is not the end of the world. People act like you should have this whole marketing system set up and have 1000 users at launch. If you have a waitlist for an app but takes more than a few months to build it, customers will lose interest. It's much easier and worry-free when you finish the app and then do your TikTok and Instagram posting.
1
17
Does anyone else have an index.ts file with 900+ lines? I've heard it should be smaller, so I'm a bit worried.
1
2
29
Working on a big 400-line plan at the moment. After it's done, my first app will be 90% ready. Launch time is getting closer; I can't wait to learn about content creation and app marketing. How close are you guys to finishing your own projects?
10
Why does nobody post their thoughts or the things they have been working on? All I see on my timeline lately is just question posts. What would you prefer/choose? Or what would you do if... It's good to have these types of tweets, but I think they are overused at this point.
1
20
How many posts per day do you guys do? I usually do 1-2 posts and 10-20 replies. How about you?
1
1
24
Harsh truth: If someone can copy your app in a week, the issue isn't building in public—it's building a thin wrapper. When you have no additional value past an API call and a simple prompt, you don't have a moat. It's always better to try and build a thick wrapper instead, meaning you add extra layers of code that add value before even touching an LLM. If you do decide to build a thin wrapper and do it in public, then it's better to keep it secretive so as to not reveal enough to get copied before launch. But even then, they would simply copy your app after it launches.
2
50
I heard about Claude's research recently. I was having a conversation about my app with my friend, and he recommended it to me for brainstorming. Do you guys use it often? Is it good?
1
2
34
I'm finally free. No idea why they flagged me in the first place. I was doing 1-2 posts and 10-20 replies per day, which isn't a big volume of posts, or at least I thought it wasn't. Has anyone else experienced such a shadow ban?
49