Pierre

Every CEO on the planet regardless of industry or company size, needs to do these 3 things: - Spend $20 to get ChatGPT Plus and experiment with the Advanced Voice feature. - Book a demo of the Apple Vision Pro at their nearest Apple Store. - Catch up on the last 12 month of robotic advancement. OpenAI Figure and Tesla Bot are a good place to start.

@chris__sev @claudeai Get better real time tracking and spot the anomalies in your tools etc => lobsterone.ai

i see this every week now on the $200 plan @claudeai give me a $500 plan. i know it's worth it

@VicVijayakumar 💯 at least they were using AI which is more than most big corp can say 😅

I used to think it was a mistake to measure AI adoption in orgs using token consumption as the proxy, but I'm coming around to thinking it's actually a way to solve the right problem with the wrong solution. Token usage leaderboards are fairly meaningless, and usage varies wildly between senior and junior engineers (juniors likely to use *more* tokens as they forceprompt the models). It's pretty close to using lines of code as a proxy for productivity. Engineering leaders at bigcos are trying to increase AI adoption and have limited tools to do so. You can only keep telling your engineers "how are you adopting AI" / "why is your estimate for your task 2 weeks and Jon's estimate is 1 day" / "why aren't your flaky CI tests fixed yet" / "why haven't we released anything new recently" over and over for so long. Tracking token usage and poking the people who are using no tokens (and also delivering slowly) is one of the ways to force innovation. The theory is if you force people to use more tokens, they'll start trying random things, become an AI-first person, and accidentally innovate — and this temporary expenditure is perfectly fine as a kickstart.

Genuine question: if every major company starts scanning their entire codebase with Claude Code (auth flows, security logic, vuln patches, all of it) isn’t that a massive privacy concern? Anthropic would theoretically get visibility into the core product logic and security infrastructure for huge chunks of the world. Even with enterprise “no training” promises and zero-retention policies, that’s a lot of trust to place in a single vendor. Am I overthinking this, or is it actually problematic?

@hemeon Track it on Lobsterone.ai :) I’m looking for a handful of VERY heavy users to see if the metrics ingestion pipeline can take it. 🚀

How much have you spent using AI? Especially those of you using frontier models? I think I'm easily at $500 a month...

@kieranklaassen @trq212 Also added a "Diagnose" function that looks at 30 days of past metrics and flags potential issues, gives tips etc.

@kieranklaassen @trq212 all done, took a few more tokens than I thought 😅 lobsterone.ai/blog/dashboard…

Resolved!! @trq212 helped me out debug where the token usage came from and it was my fault 100% Script to find token usage gist.github.com/kieranklaassen… I had a recurring script that ran every 5 minutes that should not have run every 5. I hope we can make it easier to detect these within Claude and Claude Code soon too.

@Jordanwabouzeid How do you know he’s homeless? He’s probably an atlassian cofounder or something.

The homeless guy in front of me at Sydney Library has twitter premium.

@guylikericky Space Ravens.

What form of communication are the astronauts using, aren't they 400,000 km from earth and we are getting feedback in real time

@ElectricCarAust @AutoPap Not bad! Great alternative to Tesla/China. Now I want the R5 or Twingo :)

@AutoPap Yep, and prices for EVs continue to drop. A$10k price cut from Renault on Megane eTech: electriccaradviceaustralia.com.au/EV-library/ren…

Who would have thought that Trump would be the man who accelerated the sales of EVs 😂

@hingeloss Very good point… 🤦🏻‍♂️ I wasn’t thinking straight :)

@pierre_s hmm but actually entering the website, navigating to password reset, etc, is non trivial for a script, and where you'd actually want a computer use agent

would it be a bad idea to have claude agentically roll passwords? ie go through 1password, login to each site, and update with new password?

@constantout Move to Australia and this is a lot less of an issue ;) app.lobsterone.ai/u/pierre

I feel like the whole Claude Clode limits drama is turning into a meme. I’m on $100/mo Max and have never hit limits in 2 prompts. How tf are people doing it?

@OpenAI Late April’s fool right? Right?

ChatGPT is now available in CarPlay. The voice mode you know, now available on-the-go. Rolling out to iPhone users running iOS 26.4+ where CarPlay is supported.

@IliaWhy @mattpocockuk 💯 that’s the way. I’ve been using it for a couple of weeks on work stuff and it’s great.

@mattpocockuk Use /planmode with /brainstorming from Claude’s superpowers plugin and it will ask you shit ton of questions for every step…people are missing out on this

I have also stopped using plan mode It creates a plan FAR too eagerly and usually asks you zero questions en route The whole point of planning is to get on the same wavelength with the LLM, not to generate an asset you don't read /grill-me all the way

@IamJoshJackson @lydiahallie I should add that I’m on 5x plan for 2-3 terminals and team premium plan for 3-5 more on work laptop.

@IamJoshJackson @lydiahallie I’m in Australia too (Perth) and run 4-6 terminals with CC all day / all week. I’ve hit my quota limit once last week after a big morning that’s all. Seems that the quota bug(s) are not consistent at all. app.lobsterone.ai/u/pierre

Thank you to everyone who spent time sending us feedback and reports. We've investigated and we're sorry this has been a bad experience. Here's what we found:

@official_taches @trq212 @bcherny @AnthropicAI Get proper metric tracking in place with something like lobsterone.ai - it’ll be interesting to see the volume / quota depletion.

When will we have a fix to the insane caching bug? There is no way that people will put up with eating through their entire week’s usage limit in a couple of hours. This is NOT OK for a $200/m plan guys. @trq212 @bcherny @AnthropicAI

@Shpigford @claudeai That’s mine just now

@Shpigford @claudeai Get some visibility on your quota/tokens with something like lobsterone.ai and you can see your limit in the menu bar of your Mac.

Maybe this is my issue today? My @claudeai limits were dropping an obscene amount. Usually use maybe 10-20% of my session limit but this morning it used 100% in less than an hour.