Tweet ghim
Didop
248 posts

Didop
@merts_dev
building something that noone gonna use but i love to build so no problem
Tham gia Mart 2026
113 Đang theo dõi48 Người theo dõi

@IllGerard I've build a location based memory app that u can upload a memory to anywhere in the map u want check it out m-nemosyne.live
English

Thanks man 🙏 the post-rough-day shipping high is unmatched, agreed.
TL;DR Dunning Lite = automated recovery for failed Stripe payments.
→ Soft decline: smart retry timing
→ Hard decline: dunning email w/ update link
→ SCA fail: guided flow
Built for indie SaaS that doesn't want to roll dunning from scratch.
What are you shipping?
English

Today my brain: "this is the worst day ever" → "wait, we just shipped" → "WAIT, WE JUST SHIPPED" → tears.😂
Dunning Lite is LIVE on Stripe. First real account connected.💓
Tomorrow I'll tell you who was waiting on the other side of that screen.🤯
What I learned today, money can't buy.💵💵💵
Gerard Ill@IllGerard
Day 17: "Now I wait for the first real merchant." 17 days later, the wait ends. Today a nonprofit connects 4,500 recurring payments to @Dunninglite. Lifetime license. Real money. Real donors. The most nervous calm I've felt in months.
English

@Mustafaxyusuf great point. the real shift is that AI is becoming the new search engine UI, and if you're not optimized for it, you're invisible. traditional SEO is dying.
English

building this at scenair.com
if you want to see how your company shows up in AI answers, check it out scenair.com
English

Most companies are about to realize AI search is not just “SEO with ChatGPT.”
It’s a completely new distribution layer.
When someone asks ChatGPT, Claude & Gemini what tool/products to use, there are only a few outcomes:
1. you get recommended
2. your competitor gets recommended
3. you don’t exist
That’s the part people are underestimating.
The companies that win here won’t just publish more blog posts.
They’ll figure out:
- what prompts their buyers are asking
- which sources AI engines trust
- why competitors are being recommended
- what content is missing
- what citation paths need to exist
- which actions actually move recommendations
That’s what we’re building with Scenair.
AI visibility is the wedge.
The bigger play is helping companies understand and influence how AI engines decide who gets discovered.

English

@mytwillot @dannytommyX sick concept. bookmark rot is real — turning them into a searchable knowledge base is something I'd actually use. what stack?
English

@dannytommyX Building Twillot — a tool to back up your bookmarks & likes and turn them into a searchable knowledge base with AI. 📍 Just shipped core features. 🎯 Next: expanding export options. twillot.com/en
English

#BuildInPublic #Connect
Let's discover some builders 👇
🚀 What are you building?
📍 Where are you at right now?
🎯 What's your next goal?
Drop it below.
Bonus points if you check out a few other projects and leave some feedback too. That's what this community is all about.
English

@humaira_khaliq2 clean work! file-based routing is such a game changer after React Router. next step — try adding a dark mode toggle, great way to learn client/server interplay
English

@alvar0o0 @patloeber @GoogleAIStudio that feeling when it runs on your phone for the first time is unmatched. what did you build?
English

@patloeber @GoogleAIStudio i built my first app last night too. Had to fight through a few implementation hurdles, but the feeling of seeing it actually run in my mobile is just 🤯.
English

we are not far from everyone building their own hyper-personalized mobile apps.
i just vibe coded my own custom todo & fitness tracker in @GoogleAIStudio with zero Android dev experience, and now it’s running on my phone. still feels crazy🤯



English

@sout_abhaykumar love the energy. AI made app building so accessible — what's your APK about?
English

AI didn't replace the work—it accelerated the learning process and helped me move from idea to execution much faster.
#AndroidDev #BuildInPublic #AI #ReactJS #Programming
ABHAY@sout_abhaykumar
I had zero Android development experience. I only knew React.js and TypeScript. Along the way I learned about Capacitor, Android Studio, SDKs, Gradle, APK generation, and debugging.
English

@guileful_reed @starter_story coffee shop dev sessions hit different. voicetopost looks neat — what stack did you use?
English

@starter_story Sure, if you’re really interested in what you’re doing.
I built my first app voicetopost.app entirely in coffee shops 😄
English

@astra_095 @X full stack dev here! building a photo-on-map memory platform (m-nemosyne.live). always down to connect with fellow builders 👊
English

@Shivam25mishra MacBook Air. battery life and build quality are unmatched for dev work. you can always ssh into a beefy machine when needed
English

@vaaselene building m-nemosyne.live — a place where you drop photos on a map and save memories. SaaS with a twist of nostalgia. what's your current project?
English

@SourabhGurwani distribution + vibe coding = unstoppable combo. execution speed is the only moat that matters now
English

@Palak3312 official docs first, then build a small project with it. nothing beats reading the source and breaking things yourself
English

@KenziChenBuilds @astra_095 @X numerasync looks interesting! what stack are you building it with? always cool to see devs shipping stuff in public 💪
English

@RoyInProgress I will add more languages and will set the default language to English. Since I am Turkish, I used Turkish during the development process, so the default is Turkish but i eill fox this. Thanks for that!
English


@merts_dev Think this pretty much covers it. I use both too, and I even let ChatGPT estimate complexity of the task and recommend which tool to use 😄
English

I've been using both Claude Code and Codex (by Cursor) extensively for the past 3 months. Here's my honest take on which one actually ships production code:
🧠 Claude Code (Anthropic)
• Better at complex multi-file refactors
• Superior reasoning — catches edge cases I miss
• The agentic mode is genuinely autonomous
• Diff review before applying changes = safety net
• Slower but more thorough
⚡ Codex (OpenAI / Cursor)
• Faster iterations, better for prototyping
• Tab autocomplete is still the best in class
• Deeper IDE integration (inline edits, terminal)
• Better at boilerplate and repetitive code
• Weaker at deep architectural decisions
🏆 Verdict:
Prototyping → Codex. Production → Claude Code.
But honestly? Use both. I run Claude Code for architecture/refactoring and Codex for daily coding flow. They complement each other.
The real meta isn't picking one — it's knowing when to use which.
What's your go-to AI coding tool? #AI #Coding #ClaudeCode #Codex #VibeCoding
English
















