Jan M

@DeryaTR_ @NotebookLM Is this only available in Ultra?

Here is part 2 of my plumbing training videos created by @NotebookLM. My goal is to show that it is now possible to create AI training videos on pretty much anything. These will be incredibly useful for apprentices and hobbyists. In the age of AI, you can learn and do anything!

@AnthropicAI extended the promo for another day, until 28 March. What are you building? claude-2x-tracker.pages.dev

Use the /yt-pipeline skill to generate video overviews in @NotebookLM using Claude Code and even Codex thanks to Chase AI for this workflow

What a lovely game and its vibecoded! Very nice Danny!

Are you utilizing your @claudeai usage? track it here claude-2x-tracker.pages.dev

@andrewbrown thanks for doing this. I appreciate it.

Claude Code Essentials is being packaged right now and will ship asap. The Claude Code Architect is being developed and so my target to complete is Friday The Claude Code Bootcamp (Claude Code Engineer) is happening we are just deciding on the bootcamp data. Claude Code for Zero Claude Code for Enterprise Architects up in the air for now.

There are two things that fall out side the scope of my Claude Code Courses, that I could potentially develop as paid courses. Enterprises, and Absolutely Non Technical Professionals. Let's just workshop that a bit together: > Claude Code from Zero There are professionals coming from other verticals that want to use Claude Code but they have never touch code in their life. I would have teach them ALL the steps and leave nothing out no matter how small. I would also have to guide them entire lifecycle of app development and provide full mentorship. > Claude Code Enterprise Architect Orgs want to adopt Claude Code but its not enough to just learn how to use the tool, they those workloads to be fully productionize. That means security, governance, develop onboarding, incident response and rollback. I don't mean on paper, I mean for real.

@mvanhorn @slashlast30days @Jason Do you use the Scrapecreators API? Is this much depth if you use it?

Perplexity Computer vs. Manus My Computer. I used @slashlast30days for social signal - 89 Reddit threads, 79 X posts, 30 YouTube videos, 11 HN stories. Ran just for you @Jason TL;DR: Perplexity owns the cloud. Manus just went local. Same war, opposite bets.

@mustafa_ramax @Tim_Damasc @p_naix Thanks feel free to use it as Im counting down until 27th

@Tim_Damasc @p_naix btw this one looks correct for me in Sydney, calm & well organised: claude-2x-tracker.pages.dev x.com/jannotjohnn/st…

Are you in Claude Peak Hours? Find out here claudethrottle.com

My biggest takeaways from @pmarca (co-founder of a16z & Netscape) on @FoundersPodcast such a great podcast episode For builders, too much introspection can become a drag coefficient. Andreessen says he has "zero" appetite for introspection because people who dwell on the past often get stuck there, both at work and at home. He points to Sam Walton as the archetype: Walton did not wake up wondering about his inner life, he woke up wanting to build more Walmarts. His caveat is that some elite founders are highly neurotic, so the real lesson is not that reflection is always bad, but that relentless forward motion is often more useful than self-analysis when a company is still being built. The same trait that powers a founder can make that founder hard to "fix." His psychedelics discussion is really about tradeoffs, not drugs: some high-performing founders seem driven by insecurity, dissatisfaction, or a neurotic edge that also makes them exceptional. Andreessen says he has seen people come out the other side of psychedelics happier and more peaceful, then lose interest in running the company and effectively opt out of ambition. The hard implication is that personal serenity and founder-level intensity are not always aligned. Public impact is too external to sustain a lifetime of building. Andreessen agrees with Daniel Ek that top entrepreneurs are not optimizing for happiness, but he pushes further and says even impact, money, and fame are still extrinsic rewards. His view is that the thing that still matters at 4 a.m. is usually some internal contest: becoming smarter, better informed, more capable, and harder to beat. That matters because it suggests the deepest long-term motivation is usually identity-level, not mission-statement-level. His entire worldview is basically technology versus stagnation. Andreessen argues that the world is still primitive and crude relative to what it could be, and that the real shortages are technology, information, and intelligence. In that framing, entrepreneurs are the people willing to turn better ideas into products, companies, and eventually whole industries, while everyone else mostly critiques from the sidelines. That is why he talks about founders less like financial assets and more like anti-stagnation agents. It is easier to teach a founder to manage than to teach a manager to invent. He leans on James Burnham's distinction between founder-led capitalism and managerialism, then argues the managerial model works best only when the world is stable and slow-moving. Once the environment changes quickly, professionally trained managers tend to preserve the status quo while founder-types are more likely to adapt, learn scale, and keep pushing. That is the core a16z thesis: start with the person who knows what to build, then train that person to run something large. The best relationship businesses stop being tribes of lone wolves and become platforms. Andreessen says venture in 2009 still looked like old Hollywood agencies: individual stars, internal politics, weak coordination, and very little benefit from being inside the same firm. The Michael Ovitz and CAA playbook changed that by making the whole firm available to every client, moving the staff meeting to 7 a.m., and calling clients before rivals even started their day. His point is that the middle of a relationship business eventually dies, and the winners become either tiny specialists or scaled platforms with real organizational force. Silicon Valley got bigger when startups stopped selling tools and started attacking entire industries.Andreessen contrasts the old world, where startups sold components like software, chips, or dispatch tools, with the newer world where Airbnb enters hospitality, Uber enters transportation, Tesla builds the whole car, and Facebook becomes the media company instead of selling tools to one. That shift is why venture itself had to scale: full-stack startups need more capital, broader support, and firms that can help them compete with incumbents directly. He sees AI as the clearest continuation of that trend because the winning companies are now raising billions, not tens of millions. Netscape taught him that invention and management have to be fused, not traded off. He describes Jim Clark as the force-of-nature founder and Jim Barksdale as the operator who could turn invention into schedules, systems, and a real company. Clark's instinct was to chase the newest idea; Barksdale's instinct was to keep the company from whipping itself around every day, while still preserving the inventive edge that made Netscape special. Andreessen's later worldview clearly comes from watching those two disciplines work best when they are in tension but not at war. The modern industrial superpower is direct bottleneck removal at the source of truth. Andreessen's description of Elon Musk is strikingly operational: go straight to the engineer closest to the problem, identify the single bottleneck in the production system, and fix that bottleneck now instead of letting it sit for six months. He says Musk can do roughly 120 five-minute design reviews in a day, then spend the rest of the night with the one engineer whose problem actually matters. The larger point is that elite management is not presentations or hierarchy, but technical truth-seeking plus cycle-time compression. The people who invent a technology are often bad at predicting what it will do to society. Andreessen mocks the automatic authority granted to famous AI researchers by noting that Geoffrey Hinton's politics predictably shape his AI forecasts, then uses Edison as the deeper example. Edison thought the phonograph would mainly deliver recorded sermons, while the public wanted music, which is a reminder that inventors often project their own values onto a technology's future uses. His lesson is to separate technical credibility from social prophecy. The first social response to a new technology is usually moral panic, not analysis. Andreessen runs through a long historical loop: writing, bicycles, jazz, rock, hip-hop, calculators, Walkmans, the early internet, and the old newspaper panic that bicycles would give women permanent "bicycle face" and make them unmarriageable. His claim is not that new technologies never change society, but that the press repeatedly packages novelty as civilizational collapse because fear is easier to sell than calibration. That makes "everyone is panicking" a weak argument against a technology and, in many cases, a familiar sign that something genuinely important is arriving.

@mehulmpt very nice Mehul, built the same but on top of cloudflare pages claude-2x-tracker.pages.dev

Announcing isClaude2x.com - quickly check if Claude is 2x for you or not 👉🏻 Your local timezone 👉🏻 Homepage: UI for you 👉🏻 /short API: Just a yes/no [for you/agent] 👉🏻 /json API: a full JSON object with metadata [for agent]

@p_naix Very nice, Praveen. I've also built the same on top of Cloudflare Pages since I don't want to spend a dime on a domain: claude-2x-tracker.pages.dev But yours is much much better.

claude-2x-tracker.pages.dev Great to be in NZ

@kr0der Same in New Zealand claude-2x-tracker.pages.dev

great 2 weeks to be an Australian if you use Claude Code 😂

track your Claude usage promotion here so that you can maximize @claudeai Claude Code etc. made by Claude and deployed in @Cloudflare pages claude-2x-tracker.pages.dev thanks @bcherny and team

The Algorithm (5 Commandments) At any given production meeting, there is a non-trivial chance that Elon will intone like a mantra what he calls "the algorithm." His executives sometimes move their lips and mouth the words. It was shaped by the lessons he learned during the production hell surges at the Nevada and Fremont factories. Commandment 1: Question every requirement. Each requirement should come with the name of the person who made it. You should never accept that a requirement came from a department, such as "the legal department" or "the safety department." You need to know the name of the real person who made that requirement. Then you should question it, no matter how smart that person is. Requirements from smart people are the most dangerous because people are less likely to question them. "Always do so, even if the requirement came from me." Then you need to make the requirements less dumb. Commandment 2: Delete any part or process you can. You may have to add them back later. In fact, if you do not end up adding back at least 10% of them, then you didn't delete enough. This is the most counterintuitive step it feels reckless to remove things, but the 10% add-back rule is the calibration mechanism. If everything you deleted stayed deleted, you were too conservative. Commandment 3: Simplify and optimize. This should come after step two. A common mistake is to simplify and optimize a part or a process that should never exist. Do not polish something that should have been deleted. Only simplify what survives the deletion phase. Commandment 4: Accelerate cycle time. Every process can be sped up. But only do this after you have followed the first three steps. "In the Tesla factory, I mistakenly spent a lot of time accelerating processes that I later realized should have been deleted." Speed applied to waste just produces waste faster. Commandment 5: Automate. This comes last. "The big mistake in Nevada and at Fremont was that I began by trying to automate every step. We should have waited until all the requirements had been questioned, parts and processes deleted, and the bugs were shaken out." Tesla learned this the hard way during Model 3 production when they over-automated, then had to physically saw robots out of the factory and throw them in the parking lot. Thanks to @FoundersPodcast I always listen 3x a week to remind myself

@DeryaTR_ Codex is pretty impressive

Vibe coded a game similar to Flappy Bird with some twists, using Codex 5.4 xhigh. The bird gets tired and must collect energy as it flies, while also gathering feathers to buy power-ups from the cash shop as you progress. Still an early version though😊: pixel-sky-hop.vercel.app

@JJEnglert @pencildev Thanks for sharing this JJ!

I've been using @pencildev for all of my design work and honestly, it's wild... it's like Figma and Cursor had a baby. You type what you want, AI agents build the design and you get to watch them work in real time. What I love: - Component-based designing - Agent swarms that design in parallel - Full Figma-style controls when you want them - Free to download — just use your existing OpenAI/Anthropic subscription I use it for all my AI design work now. Not affiliated, just genuinely impressed. I'm getting the founder @tomkrcha on for a podcast recording soon. What questions should I ask him?

@boringmarketer Is this gonna be recorded?

this guy is going to come show me and the vibe marketers community how he replaced an $8k/mo content strategist for $30 and runs it all from openclaw free workshop, live build, details below, won't be boring!