Sean

Last 2 weeks I've initiated 24 projects. Nothing ready to ship yet, but some of these are bangers ngl

There's something mesmerizing about watching GPT 5.6 use browser control. The way it flashes through the pages at the speed of sound, the glowing cursor zooming all over the page. This lil guy could do some damage.

@Snapchat Build your own: github.com/oso95/scroll-w…

I'm a huge fan of Specs, and I'm excited to get my hands on one. Made a concept scroller site for @Snapchat with Google Omni + Fable.

@elvissun Hilarious because I did this just today. So dangerous having cloudflare API linked

me: so here's an idea fable: buy the domain NOW

@sattyyouneed It can search the web now

nobody talks about AI hallucinating anymore. why?

@neerajjj6785 Fuck Claude, Marry Codex, Kill Cursor

developers, choose one forever

@yashmp2004 Barely replies to messages is the gravest sin. I reply to everyone, even the scammers

Vibe coding is wild. Met a 17-year-old with 6 side projects. Barely replies to messages. Pays for Claude and Codex. Falls asleep planning features. Wakes up chasing bugs. Ships updates every weekend. Checks analytics every hour. Just to see 9 active users. 😭 Three of them are his own devices.

@kunchenguid I've only used it as a subagent and I always tell Fable to be precise with instructions so as to not allow subs to spiral. It's gonna be a bummer when Fable is off subscription, it's a great orchestrator

after a few more days of using gpt 5.6 sol, i started noticing some issues - if you have good solutions, please share! 1. it uses technical jargons a lot. it's almost speaking its own language that looks like English but you can't understand it until you ask "what do you really mean by this" 2. it can over-engineer and spiral out of control. something that can be done with a few lines of changes can often result in a massive diff fixing everything in the codebase 3. it's overly conservative in terms of touching live environment, so much so that it overly relied mocks for development and validation and build things that don't really work in production these things can likely get tweaked in system prompts, but so far gpt 5.6 is the only model family that don't get these things right out of the box, and it's a bit annoying i love overall how intelligent and fast the model is though! just curious if anyone's run into similar challenges with the model and has good solutions

@omarvvvr 150 replies is peak

People ask me how I reached 3,100 followers in 6 months. They're expecting a secret. Here's the boring answer: • 8 posts a day. • ~150 replies a day. • Never missed a day. Growth isn't complicated. Staying consistent is.

@hakimuddinkika Nothing ready to share yet, but it'd be good to keep tabs on each other!

@iamspeculating Done. Share what you’re building

if you're: • shipping side projects (or full-time) • obsessed with clean UI/UX • learning by building in public • grinding on a SaaS idea • connecting with fellow founders • experimenting with AI agents • marketing your own work ...we should be friends. Reply with what you're working on. Let's grow together 🚀

@theohandsh Let's connect!

I want to connect with more founders builders vibe coders AI enthusiasts UI designers If you’re someone building with AI right now drop it in the comments and let's connect

@ConnorTalksAI Let's connect!

Looking to meet more people who are into AI, coding, startups, and building new things. If you post about tech, work on interesting projects, or just like following where AI is going, let’s connect!

@pcshipp vscode + claude/ hermes/ codex extensions is literally the answer

Is anyone still using VS Code instead of switching fully to Claude Code or Codex? Or am I the only one? 😅

@omarvvvr Most sad about Perplexity... they were crushing it for so long

Suddenly it hit me. What happened to DeepSeek? Sora? GitHub Copilot? Llama? Cursor? Perplexity? They disappeared.

@waghnakh_21 @quxiaoyin Submaxxing and tokenmaxxing are very distinct. Submaxxing is leveraging 1:10 to API cost, the more you use, the more you gain. Tokenmaxxing is the opposite

@quxiaoyin don't get me wrong but imo, tokenmaxxing is probably not productivitymaxxing

I set up my personal goal to be: I must used up all Claude Code + Codex tokens for Max subscription every day, or else I'm not productive enough.

@victor_bigfield For the love of the game

you spend 6 months building a saas. someone opens cursor on a saturday morning and ships the same thing by sunday dinner. so why are any of us still building?

@mattpocockuk Yep.. and never been easier to ship everything and sell nothing

Never been easier to feel busy and ship nothing Stay focused

Has anyone built a pitch deck generator with deep skills for researching TAM and all that jazz? I'm trying to make some nice pitch decks

@FarzaTV initially i thought it was all from the same function but yeah it makes sense to have differenet shortcuts for different intent. very nice!

Today we're shipping screen-aware dictation. First, we built a speedy speech-to-text (very fast, ~450ms). But, many products do this! So we went further. Now dictate using your screen as context. In Claude Code, it writes the prompt. In G-Mail, it replies in your voice. Demo: