We talk about folders all day, but the folders are FOR something. I want to know what... 🎖️Bonus points: comment with the single most painful manual process in your industry. The best comp entries come from exactly those answers.

"In a world full of answers, questions become valuable". I've heard Jake say this and on its face, it makes sense. I really want to understand this concept a bit deeper. For me, I take it as solutions today are a dime a dozen and with AI the speed and accessibility to answers have exploded. Its SO easy to become overwhelmed with all the new tools, updates, tactics, etc, constantly pouring in. I tend to want to ingest it all to always be in the know, always stay ahead and thorough with my approaches but it's exhausting. The focus now is to curate better questions. It's fairly freeing since now I free up all my mental capacity for determine the right questions that will lead me to the correct solution and leave everything else to the tools of today. So, I want to hear from the community. When you hear "In a world full of answers, questions become valuable", what does that mean to you?

I just realized something today, Jake's first classroom lesson isn't about Claude Code. It's a history lesson. Titled "A 2,000-Year Overnight Success." The argument: AI is not 70 years old. https://www.skool.com/cliefnotes/classroom/d7ae60cf?md=147b0e486c964ba78a70cdc1d2d40c5d I don't need to rehash it here; if you skipped it, go back to read it. It was the first weekend of April, I had found Jake's videos the week before, then discovered Skool and dove into the classroom. I made a commitment to myself to read each page, not throw them into NotebookLM for the summary. To my disappointment, I see a super long history lesson. I sink into my chair. Eventually, I get to the end. AlphaGo. I had never heard of AlphaGo before; somehow, that story had slipped past me. I queued up YouTube on my TV and sat down for a Saturday evening documentary. Fascinating! AlphaZero started tabula rasa. Blank slate, no domain-specific human knowledge. Just the rules and play against yourself. In four hours, it rediscovered centuries of chess openings, endgames, and positional theory - then kept going past what humans ever found. Kasparov called it "like discovering the secret notebooks of some great player from the past." Those moves weren't invented. They were already in the game. The truth was latent in the mathematical structure of chess. AlphaZero excavated it. That's the archaeologist move - applied to a machine. It didn't study the tradition. It played to the pattern. Jake's throughline: "The mistake is thinking these layers replace each other. They don't. They stack." In the classroom, he could have started with how to prompt or an explanation of what a harness is. Instead, he started with the source of the whole thing. Because you can't build conviction on a trend. A pattern that's held for two thousand years isn't a trend. That's proof. I've worked with a family lumber business. 125 years old. Founded 1900, delivering coal by horse and buggy. Today, it's digital marketing, performance ads, algorithms, and closed-loop lead tracking. Every generation rebuilt what the company looked like. But the fourth-generation president still says what his father said: "Young man, we're not in the lumber business. We're in the shelter business." The tools stacked. The belief didn't.

Came across Thariq Shihipar's (Claude Code lead @ Anthropic) take and it reframed how I think about AI output. The argument: we default to Markdown for AI-generated docs out of habit, but we stopped editing those docs by hand. Once the output is just read, Markdown caps its value: headers, bullets, tables that break past 30 rows. HTML unlocks SVG diagrams, inline annotations, interactive widgets, real layout. He backs it with 20 self-contained HTML files generated by Claude Code, code reviews, implementation plans, incident reports, design explorations. The catch he's honest about: HTML runs ~4–8x the tokens and 2–4x the generation time. So it's a lever, not a religion HTML for finished deliverables a human will read, Markdown for drafts, agent-to-agent handoffs, and anything in a repo. https://x.com/trq212/status/2035372716820218141 https://github.com/ThariqS/html-effectiveness

Personally. No I don't think they did, I think the researchers are discovering what I did already ! And I'm happy they are. I would love if you all could comment, share or tag Google in this though as I would love to work with them ! Video below. https://cloud.google.com/blog/products/data-analytics/how-the-open-knowledge-format-can-improve-data-sharing