01 MY ENTIRE SECOND BRAIN

@pro.glitch

56.3K followers · usually 2.7K views a post

Views

11.3K

Likes

491

Comments

246

Selling · Course / Info

Professor Glitchseen in 9 videos →https://www.askglitch.com/ ↗

Why it worked

The video effectively uses a clear, step-by-step animated explanation to break down a complex concept (building an AI-powered knowledge system). The visual metaphors and concise explanations make it easy to understand, and the focus on practical benefits like no lock-in and automated processes likely resonated with the target audience.

Summary

The video explains a system for organizing information called "Second Brain" which uses a raw folder for dumping data and a wiki folder for an AI to process and organize it. It highlights how this system allows for easy migration between AI models and provides a structured way for AI to understand and utilize your business information.

Structure

1Introduce the concept of "Second Brain" on a hard drive.
2Explain the "raw" folder for dumping data and the "wiki" folder for AI processing.
3Demonstrate how the AI links information and creates a knowledge base.
4Showcase the "log" file for tracking AI actions.
5

Original caption

How to Build second Brain My entire second brain is on a hard drive — not in ChatGPT, not in Claude, not in anybody's app. It's folders and files. I can unplug it, hand it to a completely different AI tomorrow, and it still knows everything about me and my business. The whole build: • raw/ — yours. Every note, meeting transcript, voice memo, screenshot. You dump, you don't organise. • wiki/ — the AI's. It reads raw, works out what matters, writes clean pages, and links them together. That's the part that matters — most "chat with your documents" tools just find chunks that look similar and hope. They have no idea two documents are related. Yours does, so it walks from one page to the ones it's tied to and picks up the context it was missing. • index.md — the map. Without it, the AI reads your entire brain every time you ask one question. • log.md — the memory. So a new document six weeks later updates the old page instead of duplicating it. • A hook or a schedule keeps it running without you. And here's why it's the foundation: when you build AI employees later — the one answering your email, the one going through your numbers — they don't start from zero. They feed from this. They already know how you talk, what you charge, who your customers are. Because the best model on earth still doesn't know your business. This is the thing that closes that gap. Full step-by-step build + templates this week. Check my profile. #aiagent #aiautomation #secondbrain

More from @pro.glitch

Their other posts in the index, biggest breakout first.

On-screen text

01 MY ENTIRE SECOND BRAIN // ON A HARD DRIVE IS ON THIS HARD DRIVE // WHAT IT ACTUALLY IS FOLDERS AND FILES. NO DATABASE, NO LOGIN. // AND IT COMES WITH YOU UNPLUG IT. WALK AWAY. IT IS STILL YOURS. // HARD TO A COMPLETELY DIFFERENT AI the folder doesn't care who reads it. IT STILL KNOWS EVERYTHING ABOUT ME AND MY BUSINESS 100 DAYS NOW 75% GENE-WARREN 02 LET ME BE STRAIGHT WITH YOU // THE FOUNDATION // THE SMALL VERSION OF THIS IDEA what does this document say about our refund policy? "give the AI a document and ask it questions" THAT MASSIVELY UNDERSELLS IT THE FOUNDATION EVERYTHING ELSE SITS ON IT // IF YOU BUILD ONE THING BUILD THIS BEFORE THE YEAR ENDS // EVERYTHING YOU DO WITH AI AFTER IT GETS BETTER THE MOMENT IT EXISTS 03 ONE FOLDER. CALL IT SECOND BRAIN // THE WHOLE STRUCTURE SECOND BRAIN/ or "acme plumbing" EXACTLY TWO FOLDERS. NOTHING ELSE. wiki/ BUILD THAT THOUGHT THAT ONE'S YOURS // INTO RAW. EXACTLY AS IT ARRIVES NOTES TRANSCRIBES VOICE MEMO // THE RULES FOR RAW - no naming scheme - no folders inside folders - no tidying it up, ever YOU DUMP. YOU DON'T ORGANISE. 04 SECOND ONE: WIKI. // RAW. WHAT YOU DUMPED READY ON PURPOSE. NOBODY TOUCHES IT. THAT ONE'S THE AI'S YOU NEVER TOUCH IT // ITS JOB read everything in raw work out what actually matters write clean pages into the wiki // NOT COPIES SUMMARIES 2K STORE -> 5 PAGES 05 AND THEN THE PART THAT ACTUALLY MATTERS. IT LINKS THEM. // WIKI / SUPPLIERS.MD HENSLEY - TERMS // WIKI / QUOTES.MD THE JOB WE UNDERPRICED MAR 04 JUL 11 // WIKI WRITTEN BY THE AI // WIKI WRITTEN BY THE AI MAR 04 JUL 11 JOINED FOUR MONTHS APART 06 THIS IS WHAT MAKES IT DIFFERENT // LINKS BEAT SEARCH // EVERY CHAT-WITH-YOUR-DOCUMENTS TOOL "what did we agree with hensley?" IT HAS NO IDEA THESE TWO ARE RELATED // YOUR WIKI the AI already drew the lines 24 PAGES. 63 LINKS. IT WROTE ITSELF IT DOESN'T JUST FIND ONE PAGE. IT WALKS TO THE ONES IT'S TIED TO. FIND - FOLLOW - FOLLOW AGAIN AND PICKS UP THE CONTEXT IT WAS MISSING. SIMPLICITY GAUGES. LINKS GROW. 07 NOW THE TWO FILES. // THE INDEX THAT MAKE THE WHOLE THING WORK INDEX.MD ONE PAGE a map of every page CLIENTS northline.md acme.co.md PRICING rates.md quotes-2024.md PROCESS onboarding.md DECISIONS march-pricing.md EVERY PAGE. WHAT'S IN IT. WHERE IT IS. contacts, history, what they buy renewal date, who signs day rate, retainer, discounts every quote you sent the nine steps, in order why you raised it 20% WITHOUT IT EVERY QUESTION 88 PAGES READ TO ANSWER ONE THING EVERY. SINGLE. TIME. WITH IT SAME QUESTION READS THE MAP JUMPS TO THE THREE FOLLOWS THE LINKS ANSWERS 3 PAGES TO ANSWER 88 DOES IT? 08 AND THE LOG. // THE LOG APPEND-ONLY. NEVER REWRITTEN. LOG.MD WHAT IT DID, AND WHEN 06-02 processed pricing-call.md created: pricing.rates.md 06-09 processed northline-contract.pdf created: clients/northline.md 06-18 processed 41 emails updated: clients/northline.md 07-01 processed quote-2211.pdf linked: pricing.rates.md TODAY processed northline-renewal.pdf checked index. watch found ONE LINE PER JOB. FOREVER. 6 WEEKS LATER northline-renewal.pdf // IT ALREADY KNOWS clients/northline.md EXISTS // WHAT IT DOESN'T DO clients/northline.md INSTEAD clients/northline.md UPDATED IN PLACE THE NEW THING AND THE NEW ONE POINTS AT THE OLD ONE 09 THEN YOU MAKE IT RUN WITHOUT YOU. // RUNS WITHOUT YOU TWO WAYS A HOOK EVENT raw/ ON: file lands in raw/ run: read - index - wiki wait: none IMMEDIATELY 0.0s THE SECOND YOU DROP IT IN. A SCHEDULE TIME 02:00 sweep raw/ 02:04 wiki rebuild 02:09 log written ASLEEP ASLEEP ASLEEP IT RUNS IT OVERNIGHT. EITHER. BOTH. YOU STOP FEEDING IT. IT JUST EATS. 10 IT'S ALL PLAIN FILES. // NO LOCK-IN NO DATABASE SECOND-BRAIN/ RAW/ WIKI/ INDEX.MD LOG.MD PLAIN TEXT. OPENS ANYWHERE. YOURS. the folder never flinches. CLAUDE OPENAI THE NEXT ONE ANTHROPIC THE BEST ONE GOOGLE ON MAC, LINUX + WEB // THE WHOLE MIGRATION -/second-brain/ -> you point the new one at the same folder. 11 THE PART THAT ACTUALLY MATTERS. // FIVE EMPLOYEES WHEN YOU BUILD LATER PROFESSOR GLITCH builds & runs the team NOVA content & social QUILL index, docs & calendar PROCESS YOUR EMAIL ATLAS proposals GO THROUGH YOUR NUMBERS RACK servers & web THEY FEED FROM THIS HOW YOU TALK WHAT YOU CHARGE WHO YOUR CUSTOMERS ARE WHAT YOU DECIDED BACK IN MARCH THE BEST MODEL ON EARTH STILL DOESN'T KNOW YOUR BUSINESS. KNOWS THE INTERNET ALL THAT'S THE GAP. THE MODEL SECOND BRAIN YOUR BUSINESS THIS IS WHAT CLOSES IT 12 THE FULL BUILD, THIS WEEK. // CHECK MY PROFILE REAL SCREENSHOT INSIDE THE COMMUNITY TEMPLATES AND ALL index.md + log.md Courses Course Walk through 1- Foundations 2- Automation 3- AI Automations Vibe coding Mastery Templates & Resources Premium Templates Automation Templates AI Knowledge Hub CHECK MY PROFILE // THE WHOLE PATH ZERO AI EMPLOYEES TEN'S OF HOURS, IN ORDER

Transcript

My entire second brain is on a hard drive — not in ChatGPT, not in Claude, not in anybody's app. It's folders and files. I can unplug it, hand it to a completely different AI tomorrow, and it still knows everything about me and my business. The whole build: • raw/ — yours. Every note, meeting transcript, voice memo, screenshot. You dump, you don't organise. • wiki/ — the AI's. It reads raw, works out what matters, writes clean pages, and links them together. That's the Let me be straight with you That massively undersells it The foundation Everything else sits on it If you build one thing Build this before the year ends Everything you do with AI after it gets better the moment it exists One folder. Call it second brain Or "acme plumbing" Exactly two folders. Nothing else. wiki/ Build that thought That one's yours Into raw. Exactly as it arrives Notes Transcribes Voice memo The rules for raw - no naming scheme - no folders inside folders - no tidying it up, ever You dump. You don't organise. Second one: wiki. Raw. What you dumped Ready on purpose. Nobody touches it. That one's the AI's You never touch it Its job Read everything in raw Work out what actually matters Write clean pages into the wiki Not copies Summaries 2k store -> 5 pages And then the part that actually matters. It links them. // WIKI / SUPPLIERS.MD HENSLEY - TERMS // WIKI / QUOTES.MD THE JOB WE UNDERPRICED Mar 04 Jul 11 // WIKI WRITTEN BY THE AI // WIKI WRITTEN BY THE AI Mar 04 Jul 11 Joined Four months apart This is what makes it different // LINKS BEAT SEARCH // EVERY CHAT-WITH-YOUR-DOCUMENTS TOOL "what did we agree with hensley?" It has no idea these two are related // YOUR WIKI The AI already drew the lines 24 pages. 63 links. It wrote itself It doesn't just find one page. It walks to the ones it's tied to. Find - follow - follow again And picks up the context it was missing. Simplicity gauges. Links grow. Now the two files. // THE INDEX That make the whole thing work INDEX.MD ONE PAGE A map of every page CLIENTS northline.md acme.co.md PRICING rates.md quotes-2024.md PROCESS onboarding.md DECISIONS march-pricing.md EVERY PAGE. WHAT'S IN IT. WHERE IT IS. contacts, history, what they buy renewal date, who signs day rate, retainer, discounts every quote you sent the nine steps, in order why you raised it 20% WITHOUT IT EVERY QUESTION 88 PAGES READ TO ANSWER ONE THING EVERY. SINGLE. TIME. WITH IT SAME QUESTION READS THE MAP JUMPS TO THE THREE FOLLOWS THE LINKS ANSWERS 3 PAGES TO ANSWER 88 DOES IT? And the log. // THE LOG APPEND-ONLY. NEVER REWRITTEN. LOG.MD WHAT IT DID, AND WHEN 06-02 processed pricing-call.md created: pricing.rates.md 06-09 processed northline-contract.pdf created: clients/northline.md 06-18 processed 41 emails updated: clients/northline.md 07-01 processed quote-2211.pdf linked: pricing.rates.md TODAY processed northline-renewal.pdf checked index. watch found ONE LINE PER JOB. FOREVER. 6 WEEKS LATER northline-renewal.pdf // IT ALREADY KNOWS clients/northline.md EXISTS // WHAT IT DOESN'T DO clients/northline.md INSTEAD clients/northline.md UPDATED IN PLACE THE NEW THING AND THE NEW ONE POINTS AT THE OLD ONE Then you make it run without you. // RUNS WITHOUT YOU TWO WAYS A HOOK EVENT raw/ ON: file lands in raw/ run: read - index - wiki wait: none IMMEDIATELY 0.0s THE SECOND YOU DROP IT IN. A SCHEDULE TIME 02:00 sweep raw/ 02:04 wiki rebuild 02:09 log written ASLEEP ASLEEP ASLEEP IT RUNS IT OVERNIGHT. EITHER. BOTH. YOU STOP FEEDING IT. IT JUST EATS. IT'S ALL PLAIN FILES. // NO LOCK-IN NO DATABASE SECOND-BRAIN/ RAW/ WIKI/ INDEX.MD LOG.MD PLAIN TEXT. OPENS ANYWHERE. YOURS. the folder never flinches. CLAUDE OPENAI THE NEXT ONE ANTHROPIC THE BEST ONE GOOGLE ON MAC, LINUX + WEB // THE WHOLE MIGRATION -/second-brain/ -> you point the new one at the same folder. The part that actually matters. // FIVE EMPLOYEES WHEN YOU BUILD LATER PROFESSOR GLITCH builds & runs the team NOVA content & social QUILL index, docs & calendar PROCESS YOUR EMAIL ATLAS proposals GO THROUGH YOUR NUMBERS RACK servers & web THEY FEED FROM THIS HOW YOU TALK WHAT YOU CHARGE WHO YOUR CUSTOMERS ARE WHAT YOU DECIDED BACK IN MARCH The best model on earth still doesn't know your business. KNOWS THE INTERNET ALL THAT'S THE GAP. THE MODEL SECOND BRAIN YOUR BUSINESS THIS IS WHAT CLOSES IT The full build, this week. // CHECK MY PROFILE REAL SCREENSHOT INSIDE THE COMMUNITY TEMPLATES AND ALL index.md + log.md Courses Course Walk through 1- Foundations 2- Automation 3- AI Automations Vibe coding Mastery Templates & Resources Premium Templates Automation Templates AI Knowledge Hub CHECK MY PROFILE // THE WHOLE PATH ZERO AI EMPLOYEES TEN'S OF HOURS, IN ORDER