15
10 Comments

Ask me anything about building Linkeme: the tech, the growth, the doubts

I’m Julien, founder of Linkeme.ai — an AI-powered tool that generates, illustrates, and schedules social media posts automatically for founders and small business owners.

I built Linkeme out of necessity: I had no time for social media while running my consulting business. So I built an internal tool… that slowly became a real product. Now it helps others automate their presence online in a few clicks.

Over the past months, I transitioned from consulting to full SaaS, went through two major versions, moved to AWS serverless infra, and started scaling beyond local clients in Luxembourg.

Some cool stuff under the hood:

  • Multi-LLM prompt chains (GPT-o1, Claude)

  • Visual generation via OpenAI + automated overlay engine

  • Full AWS stack: Lambda, API Gateway, DynamoDB, S3

  • Vibe-coded UI (Bolt.new + Make.com at the beginning)

And now… I’m facing the classic growth wall:

  • How to market better online?

  • What growth channels are worth my limited time?

  • Should I go all-in on Product Hunt next?

Ask me anything — whether you’re curious about the tech, the journey, or how I’m navigating growth with limited resources. Happy to be transparent and share mistakes, too.

posted to Icon for Linkeme
Linkeme
  1. 5

    What’s the single biggest bottleneck you hit when trying to turn a working MVP into consistent user growth — and how did you overcome it (if you did)?

    1. 2

      The single biggest bottleneck was staying too long in “build mode” after having a working MVP. I was focused on perfecting features, fixing edge cases, improving UX — but none of it mattered without users.

      What changed everything was shifting focus to distribution. I started talking about the product, the process, the why behind it. Once I did that, feedback started flowing, and people began engaging. I didn’t overcome the bottleneck by building more — I overcame it by letting go of perfection and showing up early and often.

    2. 1

      Thanks a lot — really appreciate it 🙌

      Delimiting the MVP was tough at first. I had to shift from “what’s technically possible” to “what actually removes pain.”

      Here’s how I approached it:

      1. I watched my own behavior.

        I built Linkeme to solve my problem: I hated spending time on social media for my consulting business. So I mapped the exact steps I was doing manually — and automated only those.

      2. I skipped everything cosmetic.

        No dashboard at first, no analytics. Just: input → content → image → publish. If it didn’t get the post live, it wasn’t MVP.

      3. User interviews > assumptions.

        Once early testers used it, I asked: “What’s the one thing you wish worked better?” If a feature came up 3+ times, I considered it essential.

      So for me, MVP = only what creates the promised result. Everything else is noise until that works reliably.

      How do you handle that tradeoff in your own projects?

  2. 3

    If you had to grow your SaaS with zero ad budget — only time and effort — where would you start today ?

    1. 2

      We’re doing this right now with Linkeme, and we’ve kept a strict $0 ad budget so far.

      What’s worked best for us:

      • Indie Hackers: A strong personal story got us over 200 visits in one day. Real conversations, real feedback.

      • Dev.to and Hashnode: Writing about the actual build process and mindset shifts brought steady visibility. Posting every 2 weeks works well.

      • LinkedIn: Honest posts about the journey (not product features) consistently outperform. It’s about trust, not traffic.

      • Product Hunt: We’re preparing carefully — aiming for strong content, backlinks, and testimonials first to make the launch count.

      What didn’t work (yet):

      • Reddit: Seems like it should work, but we haven’t cracked the formula. Most posts didn’t gain traction despite being relevant and well-written.

      • Cold email: Useful for sales conversations, but not for organic growth or awareness.

      Still experimenting with SEO and try to find micro-influencer outreach. Will share more once we know what really scales.

      Would love to hear what other founders are seeing.

  3. 2

    I actually think the app is really cool and useful, how do you delimit the MVP features, how do you decide what is essential of what is not?

  4. 1

    Is your current UI also vibe coded? That's impressive, that you were able to do production ready UI with vibe coding 🤔

  5. 1

    Hey Julien, thanks for sharing all these insights — super inspiring stuff.

    Quick question on the tech side: how does Linkeme handle user subscriptions under the hood? Are you using simply Stripe directly ?