Adam Schaible

@Jason I'm in

We started an AI founder twitter group... reply with "I'm in" if you're a founder and want to be added

@TheRealEstateG6 Not a bad point but many people who trade anything besides the S&P end up with lower overall returns - including the best fund managers.

What people don't understand about the S&P is that every single person in the country who has money is also invested in it When your money goes up 15%, so does everyone else's You gained zero relative wealth. You need to outperform the S&P if you want to actually get ahead

@alt_w_v_g The code isn't the hard part.

Investment banker sent over a software add-on last Tuesday $35M ask. 18x EBITDA. Moat slide had the usual page Proprietary data. Network effects. High switching costs. Asked the analyst to build a competitive analysis He came back 47 minutes later Not with the analysis With the product Fully functional. Built in Claude. During the meeting. No code. No team. No cap table. Told the banker we were passing He asked why Told him our 24 year old analyst just replicated 18x EBITDA when he was supposed to be taking notes He went quiet for about eleven seconds Then asked if we could send him the prompt Went home that night. Wife asked how work was. Told her we almost paid $35 million for something the analyst built in 47 minutes with Claude She didn't look up from her phone "Why would you pay for that." That is the most accurate valuation framework I've heard in three years Plz fix. Thx. Sent from my iPhone

@tanney9 @niccruzpatane While the charging infra needs to catch up, wouldn’t driver time outs limit the practical mileage to 500 miles anyway? I suppose team drivers could push further.

@niccruzpatane The long range sucks. Most diesel can do double or even triple that. 1M mile lifetime is solid, if true. Could be solid for shorter local and regional hauls. Would have benefit on noise pollution. Would love to know how the engine brake sounds bc that’s a shitty aspect 4 diesel

This is the sound of the all-electric Tesla Semi. Deliveries of the new Production Version begin this year: • Long Range model has 500 miles of range with a full payload. • 1.7 kWh per mile efficiency (average diesel semis are roughly 5–7 kWh per mile equivalent energy use). • Tri-motor powertrain with 800 kW of power (~1,073 hp), 3x the power of the average diesel semi. • The battery in the Semi is designed to last 1M miles. • Standard Range model (325 mi) has a similar turning radius as a Tesla Model 3/Y. • 0.4 drag coefficient. • Independent truckers are able to buy a Semi for use, not just fleet owners. • Semi fleet uptime is at 95% due to extremely low maintenance and reliability. • Integrated safety features in the Semi protect not just the driver but others on the road as well. • Future wireless charging. • Semi uses the same 4680 battery cells found in the Cybertruck. • Semi can power a whole refrigeration trailer or any powered unit. The technology is shared with Cybertruck Powershare.

@mattpocockuk If you have multiple teams, then 3 is probably good. For a single team, depending on the size of the project, I think 3 is too few if they are also handling production support.

Feel like in the AI age, the optimal size for a team of devs on a single decent-sized project is around 3. 1 is untenable. You can't just pause development during their holidays. 2 is OK, but still a lot of bus factor to contend with. 3 is nice and comfortable. Each day the team manages the queue of tickets for the AFK agent, discusses feature requests, architecture, reviews code, improves feedback loops, shares knowledge. Probably some devs contribute to multiple teams.

@BradGroux @AnthropicAI @claudeai @DarioAmodei @FTC rug pull...

You make this announcement BEFORE you make the change. Good riddance... what a shame @AnthropicAI and @claudeai. Customer service isn't hard, @DarioAmodei. This was a rug pool and y'all got caught. I hope the @FTC fines you.

@_catwu @bcherny Do you know when we will know if we won the lottery? :) I’ve been crossing my fingers for a long time!

Excited to see everyone at Code with Claude! What would you like to hear from us at this year's sessions?

@adamscochran @IfindRetards need you in this one.

ICE did this to 2 American’s just months ago.

@Acyn How much did you pay in taxes for 2025?

Question: I'm a waiter at a local restaurant in Queens, a full time college student who sleeps an average of four hours a night and is still thousands in debt. How is a war in a country half the world away funded by the taxes pulled from my check, helping me in any way?

@bcherny @oselcuk2002 @trq212 @AnthropicAI It’s still the right thing to do (most of the time)

@oselcuk2002 @trq212 @AnthropicAI With 1m context window, most users don’t need it anymore. If you miss it, ask claude to re-enable it in your settings.

Did anyone else notice Claude Code v2.1.81 removed "clear context and accept" from Plan Mode? That was THE recommended workflow. @bcherny @trq212 @AnthropicAI what's the new guidance? #ClaudeCode

@VictorTaelin Just buy a subscription and don’t use fast mode. Work on 2 things at once.

Sorry for posting this again, I'm still processing it: It'd cost >>> $743k per year <<< to run Opus-4.6 fast-mode nonstop Literally my company cannot afford a single person using it for daily coding. And that's a shame because the experience is truly magical. I've spent the last 2 days using it on Pi (nearly $500 gone 💀), and it was the first time I kinda got into the flow state while using an agent, because the feedback is just so fast. This is not something I ever experienced before, definitely not with GPT 5.4's own fast mode. I can't wait for this kind of super fast, super high intelligence to be available for a reasonable cost...

@trq212 You should change the 'no' response to always be #3 in CC. Often times I'm pressing 2 for yes to all, and then a single yes comes through, I press 2 when I wanted to say yes.

Your Customers' Data Just Got Smarter: What Google's New AI Embeddings Mean for Your Business appaxis.com/blog/2026/03/1…

@CMMeekaOwens You must prosecute crime to the fullest extent of the law. The city is far too lenient.

@elonmusk x should not allow posts that end with “show more” as it manipulates the engagement algorithm.

@ClaudeCodeLog Windows native installer won't update to 2.1.52 from 2.1.51

Claude Code 2.1.52 is out. CLI changes unknown yet, 91 other prompt/string changes Highlights: • Delete requests now target the specified path • POST requests now indicate when data is already archived • Initiates POST requests to the specified path Full details in thread ↓

@arunoda @JeffDean It’s getting better now.

@schaibaa @JeffDean Yeah. I can see that too. Especially for the 3-pro. This is from the direct API. I assume other labs are testing and consuming it :)

Today, we’re continuing to push the boundaries of AI with our release of Gemini 3.1 Pro. This updated model scores 77.1% on ARC-AGI-2, more than double the reasoning performance of its predecessor, Gemini 3 Pro. Check out the visible improvement in this side-by-side comparison, showing Gemini 3.1 Pro’s crisp animation built with pure code. Read more about today’s 3.1 Pro update: blog.google/innovation-and…

@sundarpichai I'm getting 503 errors on the API. This is for both Gemini 3 and Gemini 3.1. "This model is currently experiencing high demand" - but it's been happening intermittently for the past few days.

Gemini 3.1 Pro is here. Hitting 77.1% on ARC-AGI-2, it’s a step forward in core reasoning (more than 2x 3 Pro). With a more capable baseline, it’s great for super complex tasks like visualizing difficult concepts, synthesizing data into a single view, or bringing creative projects to life. We’re shipping 3.1 Pro across our consumer and developer products to bring this underlying leap in intelligence to your everyday applications right away. Rolling out now to: - Developers in preview via the Gemini API in @GoogleAIStudio - Enterprises in Vertex AI and Gemini Enterprise - Everyone through the @Geminiapp and @NotebookLM

@AWSSupport Thank you very much!

@schaibaa Hi Adam, we're sorry to hear about your case. Please check your inbox for an update from our Support team. ^GJ

@AWSSupport My case 176970655900407 has been open for several days without a response. Please help - SOS!