Dmitry

@EugBass i think next week bro

@dmtr_btc this playbook is simply overwhelming, wen part ll?

100 views. 100 views. 100,000 views. MrBeast calls that kind of explosion an "outlier". He spent years and millions building an internal system just to find the hidden patterns that make one video 1,000x bigger than the rest. Most creators post and hope. The smartest ones study why one video blows up - before they make the next one. I reverse engineered the exact same method. Without the $2 million team. I took hundreds of top videos in my niche, dumped them into NotebookLM + Claude, and let the system find the invisible patterns i had completely missed for years. The result? 212 views on video #1. 150,000+ on video #10. Same method. Different tools. If you want to build a viral YouTube channel from zero in 2026 - this is the playbook (part 1). Full step-by-step breakdown (what i did, what worked, what didn’t) 👇

@airdropdaily_3 yeah, mr beast a youtube legend

@dmtr_btc Mr. Beast definitely knows his business on YouTube 😎

@0xbobaaa harvesting panic like the house

🐋 $240,000 on Polymarket from just 86 bets. no predict, no longshots meet JnStTrdrBnusFnd. $2.8M on the table, $240,988 banked. his whole edge is selling fear to panicking crypto traders every time CT melts down that ETH is crashing to $1,600, somebody has to take the other side of that bet. he is that somebody, every time he isn't predicting price. he sells the crash the crowd is scared of, pockets the 90 cent premium, and bets the panic burns out before the deadline. it usually does each trade nets maybe 5-30%. sounds tiny. but he fires $100k+ per position, so 8% on a crash that never lands is still five figures a pop and it is not free money. look at the dip in the middle of his curve. that is fear actually spiking, his NOs going red, the account bleeding for weeks. he just sat in it and let them resolve. sitting through that is the entire job 86 bets. $240k. he stayed the house while everyone around him played scared

@0xbobaaa creative iteration at infinite scale

ONE PROMPT NOW TURNS A PRODUCT PAGE INTO 20 FINISHED AD VIDEOS WHILE YOU SLEEP most people making UGC ads still book a creator, ship the product, and wait two weeks for footage that might not even convert builder skipped every step of that by wiring Claude straight into the render pipeline. he pastes a product URL. it writes 20 different ad scripts, picks the hooks, and hands each one to an AI actor nothing gets filmed. no creator, no shoot, no shipping. every video is a real ad with a real face holding the real product, generated in an afternoon each script carries its own angle, problem-agitate, before-after, price-anchor, so the batch isn't 20 copies, it's 20 shots at a winner the part that breaks people: ad creative dies in 3 weeks. this refills it in an hour, for the cost of a coffee watch the pipeline generate a full batch below ↓

@l1ndlee data is the moat

@dmtr_btc the moat isn’t NotebookLM or Claude. it’s the labeled archive of wins and misses from the same audience - the model just makes that dataset queryable

He spent years trying to crack the code of viral YouTube. Watched thousands of videos. Filled notebooks with timestamps and patterns. Copied thumbnails, hooks, pacing - everything. Still stuck at low hundreds of views. Then he dumped his entire archive into NotebookLM + Claude. In one single pass the system found the invisible patterns he had missed for years - the exact psychological hooks, retention triggers, story rhythms and thumbnail formulas that separate dead videos from 150,000+ view monsters. It wasn’t about working harder or watching more. It was about seeing what almost no human eye can catch. Just like Kasparov felt something alien across the board from Deep Blue. Just like Paul Tudor Jones overlaid the 1929 chart and saw Black Monday coming. The best creators aren’t smarter or luckier. They built a system that sees what everyone else is blind to. Here’s exactly how i turned that same method into a complete YouTube intelligence system - and took one format from 212 views to over 150,000. Full breakdown in the article 👇

@MichealFXs that’s exact research by using NotebookLM to distill top channels into 4 core bibles plus a visual one then merging them into a Master DNA Bible Claude crafts a genuinely soulful, funny script thats light years ahead of standard AI slop

@dmtr_btc That’s the thing though, sometimes you can analyze the algorithm to death but it still feels like you’re missing the "soul" of why people actually click. Does the AI actually help with the creative spark, or just the mechanics?

@BadgerWasteland i hope you enjoyed it

@dmtr_btc 👍

@tcapy_ton thanks

@jurlycat exactly it mines 4 bibles from top channels so claude can script winning videos based on deep research

@dmtr_btc NotebookLM is a research tool, not a YouTube growth engine.

@Wagmi_forever volume, value, and brutal consistency

@dmtr_btc Is there a secret to X success? 😁

@crytonbuton remixing creates the ultimate baseline

@dmtr_btc Pattern mining is powerful, but can AI actually find originality or just remix what already worked?

@leo_marvel_ it’s a sharp analytical engine, once fed the right inputs it turns raw noise into actionable clarity fast

@dmtr_btc Thought that it's a research tool

@nahid_pro09 thanks, systems beat blind effort every time

@dmtr_btc 212 views to 150 000 is unreal what a shift

@Btcniumowang zooming right into what matters

@dmtr_btc AI is the new magnifying glass

@khodeSL data turns intuition into results that’s where the real leverage is

@dmtr_btc data analysis could change those low view counts

@FindAngus i hope you enjoyed it

@EugBass prime kasparov was truly built different that peak era was something else

@dmtr_btc old good Harry, not the same as now...

@Nekt_0 appreciate the read glad it landed well

@dmtr_btc this is a decent read

@0xbobaaa systems do the heavy lifting it works so you don't have to sink in endless hours

@dmtr_btc I wouldn't be able to set aside that much time