Profile picture of Jean-Yves Delmotte
Jean-Yves Delmotte
Co-founder @ BuddiesHR.com • 5x SaaS Founder • YC alum
Follow me
Generated by linktime
Stay updated
Subscribe to receive my future LinkedIn posts in your mailbox.

By clicking "Subscribe", you agree to receive emails from linktime.co.
You can unsubscribe at any time.

I've been a huge detractor of vibe-coding... AND BOY, WAS I WRONG!!! I tested vibe coding tools for months. Cursor was cool for shipping a quick MVP with one feature, but terrible at working in an existing code base or adding a second feature without breaking the first one. Then I tried JetBrains AI. Same story. Then I came back to Cursor. It became a bit better with existing code bases, but not enough to make me actually faster, so I kept writing the code myself. Then, 2 months ago, I tried ChatGPT Codex from the command line. And I was... SHOCKED Shocked because I had been so vocal about how bad vibe coding was. Today, I use it 80 percent of the time when I build new features inside big existing code bases. But there is a trick, you need to check 3 conditions: 1/ You must be able to review the code it generates. If you are not a software engineer, you will break feature A while shipping feature B. There is no way around this. 2/ It only shines inside an existing code base with thousands of lines. You still need to structure the project, pick the right libraries, design the architecture, and make real tradeoffs. AI cannot guess that. Once the foundation is solid, that is when the magic happens. 3/ You must prompt the how, not just the what. Tell it which files to edit, where the logic lives, how the DB schema works. You still need to know how you would build the feature yourself. AI just speeds up the execution. It takes practice. If you meet those 3 conditions, vibe coding is insane. Remove one, and your life becomes a nightmare 😅 And yes, I still think the Lovable and Replit vibe coding experiences are sh*t ^^' Agree? Disagree? Curious to get your take in the comments 👇 ---------- 👋 Hey, I'm J.Y Delmotte! I'm the co-founder of BuddiesHR, the #1 suite of Slack apps. Using Slack? Give it a try, it's a no-brainer. I post 2x a week (sometimes more) about my journey and share what I've learned along the way.  Follow me for more content like this 👆
25 comments
November 27, 2025
we grew BuddiesHR to 42k MRR while being 100% remote while most companies still believe in-person works better LOL I know what you think, remote work in a 2 people company is way easier than in a big corporation. FALSE Ask 100 founders, 99 will tell you the same thing: In the early days of a startup, nothing replaces in-person work. Not for us I guess 🤷‍♂️ (cc Fabien Pinel) So what do we do differently? 1/ Focus on remote strength We build our entire workflow around async work, we protect deep work sessions like our lives depend on it, and we invest heavily in written communication so we never depend on instant answers. When text is not enough, we record Looms so we can stay async while still giving full context. 2/ Take time during calls to talk about personal stuff Yes, it might sound off-focus, but it is everything. We start or end calls by talking about last weekend, next weekend, sport wins, family updates, whatever. Those moments matter. You cannot build a long-term company if every Zoom call is a sprint review. 3/ Organize regular team meetups Remote is great, but nothing replaces real life bonding. We run together, go to the sauna together, eat great food together. And while doing that, we talk about the next quarter, the next year, the long-term direction. People think remote makes teams weaker. For us, it made us sharper, faster, closer. What about you? ---------- 👋 Hey, I'm J.Y! I'm the co-founder of BuddiesHR, the #1 suite of Slack apps. Using Slack? Give it a try, it's a no-brainer. I post 2x a week (sometimes more) about my journey and share what I've learned along the way.  Follow me for more content like this 👆
23 comments
November 26, 2025
I launched my first startup in 2016 at age 22. of course it made $0 of revenue ^^' Why? I made every possible mistake. 1/ Launched a B2C app B2C is brutal. It is insanely hard to get someone to pay $5 with their own money. It is way easier to find a B2B buyer ready to spend $1000. If you are thinking about launching a B2C company, please do not. 2/ Did not validate the pain Like 99.99% of software engineers, I skipped validation and went straight to shipping code. And like 99.99% of them, it did not work. If this is you, read The Mom Test. It will save you years. 3/ Thought pricing was optional I lied to myself thinking I would find a business model later. The truth is I was scared to ask people to pay. The truth is simple: a product with no price is not a product. 4/ Thought ads were a real business model Whenever someone asked how I planned to make money, I answered: I will put native ads in the app. If I had $1 for every time I said that, I would be so rich. Ads are not a business model for early stage apps. Period. 5/ My ego was massive I was a young engineer dreaming of being the next Zuck. People saw me code and told me I was a genius. And I believed it. Ego is the enemy. If you cannot put it aside, you are doomed. 6/ Could not pivot I was emotionally attached to my idea. I refused to change anything. I wanted to solve THIS exact problem (that no one cared about) instead of solving something people actually needed. 7/ Did not learn from my mistakes When it failed, I did not analyze anything. I told myself « people did not get it » or « timing was not perfect ». Total delusion. If you do not learn, you repeat the same failure in a bigger costume. TAKEAWAY Do not be like me ^^' Launch a B2B SaaS after a real customer discovery process and your success rate jumps from 1 to 30%. Curious: which mistake hit you the hardest? And did you make others I did not list? ---------- 👋 Hey, I'm J.Y! I'm the co-founder of BuddiesHR, the #1 suite of Slack apps. Using Slack? Give it a try, it's a no-brainer 😏 I post 2x a week (sometimes more) about my journey and share what I've learned along the way.  Follow me for more content like this 👆
22 comments
December 4, 2025