Pascal Lindenau

4.5K posts

Pascal Lindenau

@PascalLindenau

Indie Hacker. Nomads through SEA while building Mind and Body

Taipei City, Taiwan Katılım Ağustos 2013

382 Takip Edilen1.7K Takipçiler

Sabitlenmiş Tweet

Pascal Lindenau@PascalLindenau·6 Oca

I left my job and apartment. It’s time to live. 2026 is the year to build my business, body, and mind. Just two days ago, I made my first App Store dollars too! 🥳 I’m going to post and document everything here on X. Big shoutout to @levelsio and @robj3d3 for motivating the fuck out of me. Stay tuned!

English

115

357

106.5K

Pascal Lindenau@PascalLindenau·18h

@martinrue Im with you. For me the DB is the only thing no agent ever touches. imo most people who never had anything to do with engineering just let AI do everything. These are the Hermes is a Business people mostly. Many probably don't even turn off "train with my data"

English

Martin@martinrue·18h

I don't mean to be harsh really, but I've been in engineering for over 20 years, and *never* would it be acceptable to be so irresponsible. Not to mention the fact that if the agent can DROP it, it means it can read it, and that means the agent company can read/train personal and potentially sensitive data. Blows my mind that people are happy to admit they're doing this.

English

Martin@martinrue·18h

If you find yourself in a situation where your agent even /can/ delete your prod DB, you should count yourself lucky your customers haven't yet figured out how incompetent and unqualified you are to be handling their data. And of course they're about to find out.

English

498

Pascal Lindenau@PascalLindenau·1d

Not having a 5h limit is nice, but also leads to me burning through all my token in 24h instead of 48. 🙃

Tibo@thsottiaux

Morning. The last 48 hours of Codex and ChatGPT Work have been intense! Three important updates: - Temporarily removing the 5 hour usage limit restriction for all Plus, Business and Pro plans - Rolling out changes that will make GPT 5.6 Sol more efficient across the board and that will be reflected in less usage being used so that it can take you further. Exact impact to be quantified and shared - We hit 6M active users, and are landing a usage reset in the next hour Go do things

English

154

Pascal Lindenau@PascalLindenau·2d

@anthonyriera I can’t imagine something more simple than a Apple Watch tbh and the sensors for fall and heart attacks are top notch. I can see that „longer battery life“ etc would be good, but could also lead to forgetting to load no? Good idea either way man. 🫶🏼

English

Anthony Riera@anthonyriera·2d

I would like to find a open source / open platform like Whoop or the FitBit Air so I can put it on my grandma who is now living alone and code my own fall detection / alerting on top of it! I wanted to get her an apple watch but it's too complicated. I need something dead simple

English

637

Pascal Lindenau@PascalLindenau·2d

@twoclipping Right

English

zero@twoclipping·3d

gpt 5.6 + seedance 2.0 should honestly be illegal posted a side by side comparison yesterday and half of the replies got it wrong so here's round 2: one of these girls is real. the other is the ai rebuild. same room this time, no environment tells the receipts: $400 in creator fees for the real version. the ai one costs about $2 inside the system im running: - the character ref prompts (named imperfections stack) - the video prompts + the anti ai negatives - the one line fix every gpt 5.6 script needs - the full rebuild workflow comment LEFT or RIGHT and i'll dm you the full prompting system. must be following so the dm goes through

zero@twoclipping

gpt 5.6 + seedance 2.0 is a f*cking cheat code swapped fable out of my ugc system for it and the whole thing runs 10x cheaper and faster now proof: one of these girls is a real creator. the other is the ai rebuild. same script, same visual hooks you won't clock the fake inside the system im running: - the character ref prompts (named imperfections stack) - the video prompts + the anti ai negatives - the one line fix every gpt 5.6 script needs - the full rebuild workflow reply LEFT or RIGHT and i'll dm you the answer + the full prompting system (follow so i have dm access)

English

270

247

59.8K

Pascal Lindenau@PascalLindenau·2d

@twoclipping right bro well done

English

zero@twoclipping·4d

English

753

1.1K

139.6K

Pascal Lindenau@PascalLindenau·4d

@robj3d3 Nice! Im as well only 1-2 days away from releasing my first MCP. Hope you enjoyed the sesh.

English

Rob Hallam@robj3d3·4d

You have 1 hour left. I’m going to play pickleball with friends. When I get back I’m releasing SuperX’s biggest update yet 🔥 > API, CLI, MCP > Ask SuperX (the agent) > Signals (find customers on autopilot) > Articles (write, schedule and AI thumbnails) > Feeds And much more.

Rob Hallam@robj3d3

🚨 You have 2 days to lock in SuperX at $39/mo. After that, new subscriptions will be $49/mo. The new update will include: > Ask SuperX - your agent-powered X chatbot > SuperX API, CLI, MCP and SKILL access > Signals - agents that find leads for you > For You feed - SuperX watches X and surfaces opportunities for you > Articles - the only tool that can schedule Articles on X > Contacts - build and manage lists of users > Engage - stay connected with your audience across multiple accounts And much more.

English

168

15.3K

Pascal Lindenau@PascalLindenau·4d

Claude Token reset comes in clutch as will not have many options to go out for the next 2 days. Taipei will get hit by the biggest Typhoon since 1987. Will report how it goes.

English

136

Pascal Lindenau@PascalLindenau·5d

What is happening. Grok and Facebook now apparently have opus class models?

Mark Zuckerberg@finkd

(2) Muse Spark 1.1 is strongest at agentic performance, tool use, and computer use. It does well on long-running tasks with 1M token context window, can delegate execution to sub-agents running in parallel, and is trained to use computer interfaces on desktop, mobile, or browser.

English

138

Pascal Lindenau@PascalLindenau·6d

With your X Sub you get access to Grok 4.5 Get the MCP and let Fable or Codex delegate work to it!

Theo - t3.gg@theo

Grok 4.5 is good enough that I think I have to do a video on it...

English

124

Pascal Lindenau@PascalLindenau·6d

@marcospereeira @ClaudeDevs @trq212 @bcherny We really do. Im fine if you do it when OpenAI makes their 5.6 is live post.

English

Marcos Pereira@marcospereeira·6d

@PascalLindenau @ClaudeDevs @trq212 @bcherny come on bro I never ask for anything we really need this reset cmon

English

Pascal Lindenau@PascalLindenau·6d

Got my rate reset this Monday. Had a all day marathon and used up 80% of my weekly in 1 day. Now it gets extended till 1 day before the reset. @ClaudeDevs pls reset. 👉👈

GIF

Claude@claudeai

We're extending access to Claude Fable 5 on all paid plans through July 12.

English

353

Pascal Lindenau@PascalLindenau·7 Tem

@OpenAIDevs @stevie_builds

QAM

OpenAI Developers@OpenAIDevs·7 Tem

GPT-Realtime-2.1-mini is now available in the API, bringing reasoning and tool use to our Realtime mini lineup at the same cost as GPT-Realtime-mini.

English

125

238

3.8K

419.2K

Pascal Lindenau@PascalLindenau·7 Tem

@stevie_builds Insanity rn. I would probably grab a m4 or m3 with more ram rather than a m5 with less.

English

stevie@stevie_builds·7 Tem

This is getting absurd. I pray that I don't end up forced to use windows. on one hand: - it is the only tool i need (not like when i did carpentry). + codex sub. - it will last over 5 years - i wont have the constant battle with specs that my air has right now. - i have an $8k tax return coming on the other: - 5.7k aud is mental - in one year ill be baited into upgrading and complaining about the price.

English

Pascal Lindenau@PascalLindenau·7 Tem

@yannschaub @arvidkahl ill get the scream sound pack and try it!

English

yann 🧑‍🚀@yannschaub·6 Tem

@PascalLindenau @arvidkahl instagram.com/p/DaZpzZ4h7XM/ try this

English

Arvid Kahl@arvidkahl·6 Tem

Everyone out here fablemaxxing?

English

2.9K

Pascal Lindenau@PascalLindenau·7 Tem

@dalu_hey If I work to the fullest I can hit my 5h max in roughly 2.5h on 20x I don't wanna know how bad it is on lower tiers.

English

dalu@dalu_hey·6 Tem

@PascalLindenau I'm not even using Fable. Limits are too low. I don't want to bother with hitting 5h limits in 7 minutes

English

Pascal Lindenau@PascalLindenau·5 Tem

Have we all forgotten this?

English

1.1K

315.8K

Pascal Lindenau@PascalLindenau·6 Tem

Human progress over the coming decade will be unlike anything we have ever experienced. Buckle up.

kache@yacineMTB

I hope you guys are ready for technological progress like you couldn't believe. We're going to see some spectacular things. You should help make them

English

172

Pascal Lindenau@PascalLindenau·6 Tem

@nic_amadio That's because life is really nice in Belgium on this paycheck.

English

Nicola Amadio@nic_amadio·6 Tem

Tech workers in Belgium 🇧🇪 pay a shitton of taxes but are all somehow happy with what they get for their tax money. I mean not all but many.

English

6.1K

Pascal Lindenau@PascalLindenau·6 Tem

@amiee36120 @JCodesMore if 5.6 sol plays in the ballpark of fable I will try the blue Side.

English

114

Amiee@amiee36120·6 Tem

@JCodesMore @PascalLindenau What is the point of using Claude? I have a token usage limit of nearly 600 million in Codex, whereas standard Claude is at 200 million. So they’ll be removing Fable 5 soon, while 5.6 Sol Ultra is coming to Codex.

English

112

Keşfet

@martinrue @anthonyriera @twoclipping @robj3d3 @marcospereeira @ClaudeDevs @trq212 @bcherny