Loomin
AI ghostwriter that writes LinkedIn posts in your voice
I built a LinkedIn AI ghostwriter as a vibe coding experiment — here's what happened
A few months ago I hit a paywall on a calculator app I used daily for my manufacturing business. So I decided to build a similar tool in-house. I tried a bunch of AI tools to help me and eventually Claude one-shotted my query. I'll be honest I was genuinely surprised. I had no idea AI was this capable, even for a regular person with no coding background.
That moment got me hooked. I started experimenting more, and last month I decided to vibe code an actual SaaS product.
I chose LinkedIn content because it's a problem I see firsthand. We're a small manufacturing company LinkedIn presence and audience building are nowhere near the top of our priority list, and we don't have the manpower to delegate it either. I figured if we had this problem, plenty of others did too.
Yes, tools like Taplio, AuthoredUp, and Supergrow already exist. I knew that going in. But my main goal was to experiment with vibe coding and experience building a real product end to end the brand, the marketing, all of it. So I built it anyway.
I won't pretend Loomin is better than those tools right now. But I'm willing to make it better — and that's actually why I'm posting here. If you try it and have feedback, I genuinely want to hear it.
What I learned building it:
Vibe coding is real but you still need to understand what you're building
Distribution is 10x harder than the product
Creating a brand from scratch is its own full-time job
If anyone wants to check it out: https://www.loomin.pro/
Happy to answer any questions or feedback about what I'd do differently.
21 Comments
For sure finding users because building as you mention with the AI become much simpler.
I've developed a free, AI-based launch planner for local service businesses (cleaning, lawn care, pressure washing).
"Distribution is 10x harder than building the product" has been the biggest lesson for me too. Building is addictive because you get immediate progress. Marketing is slower and full of uncertainty, but it's the part that actually determines whether anyone uses what you built.
You already named your edge and then walked past it: you're a manufacturing company that won't prioritize LinkedIn. Build Loomin for that exact buyer (industrial and trades SMBs who hate posting) instead of going head-to-head with Taplio for general creators, and distribution stops being 10x harder because you are the ICP. I run a tool in this category, and the winners aren't the best writers, they're the ones who own a niche nobody big bothers to serve.
Thanks a lot for the advice, will work on that.
The fantastic world of AI continues to amaze us every day. With so many tools available today, everything becomes easier. With the sheer number of tools that exist and their ever-growing number, the problem now will be finding the needles we need in the haystack. For those who have already created some of the needles, the challenge is making them clearly visible. Congratulations and best of luck on the journey ahead.
Thank you!
"Distribution is 10x harder than the product" hit me in the bones. Solo founder here building voice & WhatsApp agents, and the product was never the bottleneck either. One thing that's helped, since you're honest that you're not differentiated on features yet: the distribution has to be you, not the tool. Nobody switches from Taplio for a feature, but they'll root for a builder who's open about where they're at. This post is the distribution.
"Distribution is 10x harder than the product" — this hit hard. I'm in the exact same boat with WhaleTrack right now. The building part was the easy bit, getting eyes on it is the real work. Respect for being honest about the existing competition and building anyway — that's the right mindset.
The distribution is 10x harder line is the most honest thing on this page. Most vibe coders figure that out after launch, not before. Curious what you’ve tried so far are you going after LinkedIn users directly or trying broader channels?
I am going on a more direct approach for now.
Direct is the right call early on broad channels waste time when you don’t know what converts yet. What does direct look like for you right now cold DMs, communities, something else?
The line that stood out to me was "distribution is 10x harder than the product." I think a lot of people discover that only after they ship. Building proves you can create something. Distribution proves you've chosen a problem people care enough to solve. Those are two completely different challenges, and the second one usually forces the biggest product decisions.
Exactly and I am a big believer on learning things on the road especially on that project because it is just a first step for me to go in to new enviroment I am really greatfull how much things I am able to learn in process.
That's exactly what I was getting at.
Your reply made me think about one implication that could end up changing how you interpret what the market is actually teaching you, but it's probably too much to unpack properly in a thread.
Happy to explain what I mean if it's useful. What's the best email to reach you on?
No-code founder here too. The candor on failure is valuable — most won't admit when a feature flopped. Question: at what point did you realize you needed to pivot vs. just push harder? I've seen founders spin on the wrong idea for 18 months because quitting felt like giving up.
I don't think anything has failed yet since app went online just 3 days ago.
Fair point — 3 days is still validation phase. What I'd watch for: when your first users hit a friction point (and they will), that's the moment to decide. Don't wait for a full feature flop. Even small friction signals early.
This comment was deleted 2 months ago