Luqmaan

Still climbing 📈

AI youtube is eerily similar to peak bull-run crypto youtube

@om_patel5 100%. It never reliably follows instructions in claude.md / memory..

THIS 17 YEAR SOFTWARE ENGINEER JUST WENT OFF ON CLAUDE CODE AND CALLED IT COMPLETE GARBAGE this guy has been writing code since 2008. started using AI in march. 5x max plan since day one at first he was blown away. 3-4 sessions running in parallel in tmux all day. commits in seconds. answers in seconds. implementations in minutes. barely ever hit the limit then opus 4.7 dropped and everything broke before vs after: > "commit this" used to take seconds. now takes 30 seconds > "implement this plan" used to take minutes. now takes 45 minutes > resizing the terminal used to rewrap text cleanly. now lines get interlaced and diffs get cut off > ctrl+o used to show the full thought process. now it shows nothing useful > it now says "almost done thinking" like that helps anyone > he hits the limit constantly even after learning to split opus for planning and sonnet for edits but the WORST part is that it ignores instructions he told claude a thousand times to use short timeouts (10-15 seconds) for network programming. wrote it to memory. within one implementation cycle claude switches back to 30s, 60s, 5 minutes told claude to "NEVER AUTO COMMIT" a billion times. after a while claude starts auto committing again the guy uses /caveman mode for brevity. claude forgets about it and dumps a wall of text after every implementation claude can't even follow a simple 200 line plan anymore the plan said "change the signature of handle_input to take Bytes instead of &[u8]." claude added a completely separate function instead. said it wanted to "keep backwards compatibility." on a brand new project. there is no backwards compatibility then it started deviating from plans mid-implementation. tried to cut features. wanted to "ship v1 now, do the rest in v2." there is no v1 or v2. that's the developer's job to decide, not the AI's we had something good with opus 4.6. now it's garbage

My vibe coded app just got its first subscription

@iPritamX You can setup a US LLC and use something like Mercury for payments

You just don’t know how difficult it actually is I’m building iOS apps from India, targeting the US & EU markets. The App Store gives you zero local boost because you’re not based there. No algorithmic love, no organic visibility. You’re already starting 90% behind before you even launch. Then you try to catch up with Apple Search Ads… RBI policies + Apple straight-up rejecting Indian cards make it almost impossible. Now you’re not just behind, you’re completely invisible while others race ahead. This is the brutal reality for Indian devs chasing global success.

1/ I shipped Rind in February. It took me 300+ hours from start to finish. 2/ I built the whole thing mainly to learn whether someone who doesn’t know how to code can now use AI to build and ship a real app. 3/ The answer is yes. That part is real. 4/ I also wanted to see whether an audience asset I already owned could help get it off the ground. 5/ I tried using a recipe blog I own with 100k monthly visitors as the first distribution channel. The results have been underwhelming: a handful of users each week and 0 paid subscriptions. 6/ The obvious next step would be to test more standard app distribution channels like creator/influencer content or paid acquisition. 7/ I’m just not sure yet whether I want to put more bandwidth into that part. 8/ But as an experiment, this was still valuable. AI made it possible for me to go from non-coder to shipped app. 9/ For anyone still thinking about building: the barrier to building is lower than ever. The main challenge is getting people to care

App mostly built. Just need to clean up the code a little and get it production ready. Still haven't decided on a name 😆

@david_attisaas Copycat would be when you deliberately use a similar name and branding right? Same features but completely different branding is fine?

the issue with copycat a copycat is... ...once apple cracks down on the main one all the copycats of said copycat ends up with their app removed from the app store + probably apple developer account banned cascade of bans tldr: don't copycat a copycat 😂 tldr 2: don't copycat without thinking doing a 1:1 copy is extremely dangerous I recommended to do it in the past and I was wrong it's ok to take a great inspiration for a different niche etc etc but doing a 1:1 copy will get you banned at some point because the "original" app makers will get an army reporting your app to apple and you might not get the money you made with the app

@IntrovertGeekUK Nice!

Pinterest support CONFIRMED it was a glitch. We live to fight another day...

Claude: "The profile header looks beautiful and matches the design system! Ready for the AccountSection next." The profile header:

@svpino How would you go about systematically learning the part other than just writing the code in the most efficient way possible?

@luqmaansa It helps immensely.

Building software is only 10% about writing code, and 90% about thinking what code to write, why you should write it, and how to do it well. You can delegate the code-writing part to an LLM, but so far, they have shown they can’t handle the thinking part. It’s called “vibe-coding” and not “vibe-thinking” for a reason.

Added an import progress dialog

Built a separate test UI to compare outputs between different models. Leaning towards grok 4 fast as primary. gpt4.1 nano is looking good too

Updated the time UI

I've been mostly using UX Pilot for UI because it generally did better than other tools. But for the time display and difficulty, I wasn't happy with UXP's design (Right) so I tried redesigning with vanilla chatgpt (left) and it looks way better imo

@david_attisaas Are you manually removing the watermarks?

not fucking around I have 2 guys doing this full-time for me + I'm doing it myself a few hours a day to learn what sora can and can't do

Anyone using openrouter in prod app for AI features? Or do you just bake in a few fixed models to choose from?

Recipe importing and normalizing is working surprisingly well 100% of the time. Small sample size of 100+ urls but still expected it to break more often. Now need to test several models and find fastest/cheapest option and add 1 or 2 fallback options.

I usually do 2 to 3 hour sessions and never ran into limits. Been hitting limits every hour since yesterday

@HuuVanTran @mediavine @ezoic Great stuff!

[Income Report, Sept 2025📊📉📈] In Sept 2025, our team (+ me) made $12,054.45 in revenue from 325,000+ sessions, that’s a -27% traffic drop compared to August. Thanks to higher RPMs from @mediavine and especially @ezoic, we still ended up making a bit more than last month, with Ezoic RPMs jumped from $23+ to $40+. 🌟 Fandom site (Facebook + Mediavine): Our sessions dropped hard (-45% traffic) even though FB pages' reach only fell -21%. Thanks to higher year-end MV RPMs, our revenue wasn’t too bad (around - $2K only). Our traffic dropped for a few reasons: - Team’s performance last month was inconsistent, quite bad. - Facebook recent changes sometimes hide our link comments, which cut clicks to the site even further. 👉 We blame ourselves first for the drop. Second, we’ll be adjusting to Facebook’s changes going forward. 🏡 Decor site (Pinterest + Ezoic) Traffic is steady at ~110K sessions/month for over 6 months 🥹. Our RPM this month were strong (thank you, Ezoic). Still, I’m questioning whether our real image pins are weaker than our AI image pins from before. We switched to real images about 6 months ago, right when traffic plateaued. Coincidence? Not sure yet. 🤔 👨‍🔬 Tech site (Bing, Pinterest + MV) Stable at $1.7K+ last month. This site was once MV Pro but got wrecked by HCU (familiar right?). For now, I’m letting it run while keeping a few ideas in mind:👇 - Launching a tech FB page to funnel traffic back in. - Using @blog_to_pin to automate pinning on its Pinterest account. Would love your thoughts if those are good ideas? 😉 🤏 Smaller sites These are the backburner ones left after HCU. They’re too small for now, but all are in Mediavine, so I’m happy to let them sleep until having better ideas for them. Also: +$70 from our YouTube channel (after the 30% US tax cut). 🛒 Affiliate column This is where luck came in. We only reached $9,500+ sell revenue (below the $9,750 requirement for the $295 flat fee). But Amazon Affiliate still gave us the $295 reward anyway 😍. That boosted affiliate income from +$220 last month → $675 this month. 📊 Overall This month was a traffic + feelings rollercoaster. 🤯 Yet our revenue ended at $12,054.45. Not bad right. For this, I'm thankful. I paid my team $3,475.63 (after bonuses). Which leaves me $8,578.82 profit. Thanks for reading this far, it means a lot to me🙏.