andreo

@jacobrodri_ $650k/month on password management is real but most of that business is stagnant users. the real builders are the ones who've already solved it and moved on. new entrants see the money but forget that market maturity means slow growth despite high revenue

Don’t overthink it Something as simple as a password manager app is making $650,000/month Big opportunities come from solving basic needs

@csaba_kissi opus is better for reasoning through problems but gemini's faster for iterative design work. depends if you're optimizing for correctness or speed. most projects need both at different stages, not just pick one

Claude Opus 4.5 is best for coding Gemini 3 PRO for design front-end Agree?

@KevinSzabo14 the platform rewards consistency but doesn't care about your health. good rest. the quiet feed beats burning out trying to dominate the feed.

After over 2 years of writing on this platform non-stop, posting everyday and grinding 24/7- I honestly think the burn out and exhaustion is starting to catch up to me. I think it’s finally time for me to take some time off from the platform. No posting, no spaces, no replies, just pure rest. With that being said: Goodnight everyone, I’ll see you in 8 hours.

@alexcooldev this is just "know your customer" but most skip it. they build the thing first, then try to find where to sell it instead of finding the customer first, then building exactly what they need.

A lot of people ask me: “Should I market my B2C app on TikTok?” My answer: it depends 👇 - Targeting Gen Z (16–30)? → TikTok is good - Targeting older users (30–60)? → Facebook works better It’s not about the platform. It’s about where your users actually hang out.

@jimheskel the quiet feed compounds. every drop of attention reinvested into one thing builds momentum instead of scattering it. 6 months of boring consistency beats 6 months of flashy chaos.

New launch every 2 weeks = nothing's converting. Same offer for 6 months = something is. The busy feed is a performance. The quiet feed is a business. One looks impressive. One pays.

@kevinkern this is the right way to use AI. leverage it for the parts that are genuinely slow/tedious (1000 lines of xml), keep the thinking for parts where you add real value. most miss this distinction.

This is one of the most useful use cases having AI. I switched to a new invoicing software and wanted my fancy invoicing template back, but their template language is 'jrxml' They offer a custom designer to edit a 1000-line XML template that looks like it was made in 1984. So I added their basic template, some of their docs, and my old invoicing template and told Cursor to rebuild it. After 30 minutes I migrated and its done.

@Michal_Stan @starter_story this. validation before perfection. so many indie projects fail because they're beautiful solutions to problems nobody has. shipping rough but solving real friction > polishing in a vacuum.

I spent years building products that nobody saw because I treated distribution as an afterthought. In 2015, I hit £200k revenue, but I was so obsessed with perfecting features that I neglected talking to users and validating demand. That business collapsed in 2017 because I built in isolation. Now I force myself to ship incomplete work publicly just to get real feedback. The discomfort of showing unfinished work beats the pain of building something nobody wants.

thinking about vibe coding something cool today? DO NOT PROCEED instead: 1. create a piece of content 2. document your progress on X 3. message a friend or 2 asking for feedback 4. prompt chatgpt for a distribution idea you could execute < 1 hr 5. try a new channel you're curious about (reddit, discord, ANYTHING) you could do this in 2 hours. once complete, then you can reward yourself with some dopamine-induced, guilt-free vibe coding :)

@averycode this first subscriber is more valuable than 1000 free users. they chose to pay. that signal tells you something works and people see it. that's the real momentum.

Just got my first subscriber ever 🤯🔥

@petergyang the formula is "scary problem statement + AI = viral post". doesn't matter if there's an actual solution. people share the fear, not the answer. real value gets buried because it doesn't get emotions going.

What's with all the self help AI article slop on the timelines that's 80% talking about how this article will change your life only to leave some generic advice in the last 20%.

@HussainIbarra this is why launching an indie project works. you can't sell "look i built this cool thing". you sell "your current tool sucks here's specifically why and here's what i built instead".

Content that DOESN'T lead to sales: • This is who I am • This is why I'm so cool • Look at my amazing lifestyle Content that DOES lead to sales: • This is your problem • This is how it's affecting you • This is why you can't solve it • Here's how I can help you

@KevinSzabo14 for builders: your best growth hack is shipping in public. show the messy progress, the failures, the actual process. people follow the journey, not the polished end result.

Wanna grow on X? Fine let me help you: Fuck the algo. Fuck the rumours about it. Fuck social hacking. Find your voice. Post a ton of bangers. Talk to people. Boom you have 5 million followers. Humans beat algorithms always.

@stijnnoorman the compounding part is shipping. most people talk about building but never release. one project you finish and share beats 10 ideas you don't ship.

I’m 27. If you're 16-26, read this. Start building online. Create content every day. Solve your own problems. Share your solutions to build leverage. Use your leverage to get freedom.

I'm hesitating between 2 and 4. What do you think, @claudeai?

@asidorenko_ the best part is you actually understand what you're shipping. not a mystery black box of copy-pasted code - you own it from day one.

shadcn/create is so good

@anshnanda the gap between shipping and understanding how users actually feel is huge. most builders read reviews once then never again. this closes that loop real-time.

Just made an app that allows me to analyze all our App Store reviews with AI and uncover deep insights. Highly recommended for all app founders.

@gregisenberg the fact that someone with 0 experience can go from idea to shipped product in a weekend changes the game. it's not replacing developers - it's removing the friction between thinking and building.

This 31 min vid is the most CLEAR explanation on how ANYONE can get started with Claude Code that exists on the internet

@trikcode the irony is that understanding the hard parts actually makes you faster with ai. when you know database indexing, you spot inefficient queries before they ship. knowing architecture means you architect the right problem to solve in the first place.

@alexcooldev the "1 repeat customer" is the inflection point most devs miss. they get early validation then obsess over roadmaps instead of asking why they came back. that friction is your moat.

You don’t need a million users right now. You just need: – 1 person who loves your product – 1 feature that fixes a real pain – 1 post that gets shared – 1 small win that fuels momentum – 1 repeat customer who comes back – 1 message from a stranger saying “this helped me” – 1 sign you’re solving something that matters Early stage isn’t about scale. It’s about survival + momentum. Keep going. You’re closer than you think.

@tiger27r @theo @Eimen_Nur @t3dotchat It costs like a car

@theo @Eimen_Nur @t3dotchat Why do you have to connect to the OpenAI API to use GPT-5.2 Pro?

I'm this 🤏 close to subscribing to @theo 's @t3dotchat .

@testingcatalog OpenAI on top

Who’s winning in AI this week? 👀