Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Anthropic built a model that turns any object into a 3D schematic and then runs experiments on it before anyone touches the real thing. The geometry was never the hard part. The hard part is a model that knows which parts of the object are allowed to move. >...

0xWast3's profile picture

wast3

7,100 subscribers

101,950 görüntüleme • 1 ay önce •via X (Twitter)

Haberler & PolitikaEğitimBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

11 Yorum

Gipp 🦅 profil fotoğrafı
Gipp 🦅1 ay önce

swap step in shape actually sped up my teardown workflow big time

kiosa profil fotoğrafı
kiosa1 ay önce

What a banger mascot mate

monokern profil fotoğrafı
monokern1 ay önce

does it work from a photo or does it need a full description?

Fokki profil fotoğrafı
Fokki1 ay önce

terminal is next level

beamnxw ./ profil fotoğrafı
beamnxw ./1 ay önce

how accurate are the inferred constraints from the initial description?

shmidt profil fotoğrafı
shmidt1 ay önce

good boy

Aidiots profil fotoğrafı
Aidiots1 ay önce

So cuute ...

Ridark profil fotoğrafı
Ridark1 ay önce

You have a unique terminal with a really interesting approach for real honestly for sure honestly, how did you build it out for real?

Insomnia profil fotoğrafı
Insomnia1 ay önce

bro I trying to build same

ຸ profil fotoğrafı
ຸ1 ay önce

sounds and looks like a vibecoded gigayap

swapcherry profil fotoğrafı
swapcherry1 ay önce

the swap step carrying the whole thing

Benzer Videolar

Anthropic turned one GitHub request into a living 3D model that answers every new issue with a rebuilt version of itself Opus 5.5 builds the whole system from that first request, from geometry and rig to the viewer and the pipeline behind it. > REQUEST - one issue describes the object in plain words, nothing else needed > BUILD - mesh, rig and materials generated and committed as editable source files > VIEW - a browser viewer added to the repo so anyone can spin the model instantly > LISTEN - every new issue is read as a change request against the current model > PATCH - only what changed gets rebuilt, then a pull request opens with a fresh render A 3D asset is normally a dead file. Someone exports it once, and every later change means opening the software again. Here the model stays alive inside the repo. Ask for a longer handle or a softer material, and it answers in a pull request. Each pull request carries a new render, so the change gets reviewed by eye before anyone merges it. Because the source is code and not a binary blob, every version is diffable. You see exactly which part moved and why. Rejected changes cost nothing. Close the pull request and the model stays exactly as it was. Teams already run software this way. Running a 3D model like a codebase is the part nobody had wired up. The first request, the repo layout and one full issue-to-render cycle are pinned in the article.
0:36

Anthropic turned one GitHub request into a living 3D model that answers every new issue with a rebuilt version of itself Opus 5.5 builds the whole system from that first request, from geometry and rig to the viewer and the pipeline behind it. > REQUEST - one issue describes the object in plain words, nothing else needed > BUILD - mesh, rig and materials generated and committed as editable source files > VIEW - a browser viewer added to the repo so anyone can spin the model instantly > LISTEN - every new issue is read as a change request against the current model > PATCH - only what changed gets rebuilt, then a pull request opens with a fresh render A 3D asset is normally a dead file. Someone exports it once, and every later change means opening the software again. Here the model stays alive inside the repo. Ask for a longer handle or a softer material, and it answers in a pull request. Each pull request carries a new render, so the change gets reviewed by eye before anyone merges it. Because the source is code and not a binary blob, every version is diffable. You see exactly which part moved and why. Rejected changes cost nothing. Close the pull request and the model stays exactly as it was. Teams already run software this way. Running a 3D model like a codebase is the part nobody had wired up. The first request, the repo layout and one full issue-to-render cycle are pinned in the article.

wast3

16,350 görüntüleme • 28 gün önce

Workflows vs. Graphs, clearly explained! workflows are great, and almost everyone has one. here is the ceiling: a workflow decides every step before it runs. you drew eight boxes in March. six months later the same three fire, every single time, and the other five have never once been reached. then a case arrives that nobody drew, and it goes to the closest wrong box. quietly, with a green status, because from the inside that looks exactly like success. Graph engineering fixes this by moving the decision: not what the steps do, but when the steps get chosen. you need both, and here is the sentence that resolves the whole confusion: a workflow decides the steps before it runs. a graph decides them while it runs. ↳ drawn in advance: the boxes, the branches, the order, the error path ↳ decided at runtime: how many units exist, what each one is allowed to see, which ones get created at all Prompts → Context → Harness → Loops → Graphs branches do not make it a graph. the branches were drawn in advance too, which means every one of them is a case you already thought of. the trick is knowing which part is allowed to be fixed. the node kinds are fixed. a splitter is a splitter, a gate is a gate, a merge is code. what is not fixed is how many of them exist this run, and that is decided after something has been read. one thing to know before you scale it. a workflow fails in a way that never pages anyone. ↳ the wrong branch ran, every check inside it passed, and the output is well formed ↳ nothing errored, because routing to the wrong box is not an error, it is a route that last one catches careful people. you cannot test your way out of it either, because the test suite was written from the same diagram that has the gap in it. and the one that eats whole nights: a workflow that has never surprised you is not stable, it is narrow. if it has run four hundred times and produced the same three shapes, it is not handling your work. it is handling the part of your work that fits it, and you have quietly stopped sending it the rest. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this and read it below ↓
0:51

Workflows vs. Graphs, clearly explained! workflows are great, and almost everyone has one. here is the ceiling: a workflow decides every step before it runs. you drew eight boxes in March. six months later the same three fire, every single time, and the other five have never once been reached. then a case arrives that nobody drew, and it goes to the closest wrong box. quietly, with a green status, because from the inside that looks exactly like success. Graph engineering fixes this by moving the decision: not what the steps do, but when the steps get chosen. you need both, and here is the sentence that resolves the whole confusion: a workflow decides the steps before it runs. a graph decides them while it runs. ↳ drawn in advance: the boxes, the branches, the order, the error path ↳ decided at runtime: how many units exist, what each one is allowed to see, which ones get created at all Prompts → Context → Harness → Loops → Graphs branches do not make it a graph. the branches were drawn in advance too, which means every one of them is a case you already thought of. the trick is knowing which part is allowed to be fixed. the node kinds are fixed. a splitter is a splitter, a gate is a gate, a merge is code. what is not fixed is how many of them exist this run, and that is decided after something has been read. one thing to know before you scale it. a workflow fails in a way that never pages anyone. ↳ the wrong branch ran, every check inside it passed, and the output is well formed ↳ nothing errored, because routing to the wrong box is not an error, it is a route that last one catches careful people. you cannot test your way out of it either, because the test suite was written from the same diagram that has the gap in it. and the one that eats whole nights: a workflow that has never surprised you is not stable, it is narrow. if it has run four hundred times and produced the same three shapes, it is not handling your work. it is handling the part of your work that fits it, and you have quietly stopped sending it the rest. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this and read it below ↓

Hanako

16,810 görüntüleme • 1 ay önce

🚨FROM METAPOD TO A SPIKY GLOBE At the start of this footage, the object appears to be in the form of a Metapod. It stays that way for several seconds before the camera pulls back and loses the detail. When the object comes back into view it appears wider. By the end it looks circular, with short spikes around its edge. It could be just one single object rotating through different angles, the video never really gives us a clear view of the actual change itself. We begin with a Metapod and finish with what looks like a spiky ball. #ufotwitter #uapX #UFO #UAP #Metapod #UFOSighting
0:46

🚨FROM METAPOD TO A SPIKY GLOBE At the start of this footage, the object appears to be in the form of a Metapod. It stays that way for several seconds before the camera pulls back and loses the detail. When the object comes back into view it appears wider. By the end it looks circular, with short spikes around its edge. It could be just one single object rotating through different angles, the video never really gives us a clear view of the actual change itself. We begin with a Metapod and finish with what looks like a spiky ball. #ufotwitter #uapX #UFO #UAP #Metapod #UFOSighting

Skywatch Signal

13,186 görüntüleme • 1 ay önce

Everyone who chased $TAO at 276 is now watching it in the red and wondering if the move is over. It is not. It is doing the one thing a real breakout is supposed to do. Price ran to 276, then pulled back into 252. That is not weakness. That is the retest. The level that has to hold: 252 That is the 0.382 of this whole leg, and price is sitting right on it. Hold this zone on daily closes and the pullback is just the trend catching its breath. The line under it that still keeps bulls in control: 238 The 0.618 golden pocket. As long as dips get bought between here and there, the higher low structure stays intact and the uptrend is fine. Where it goes if the retest holds: 276, then open air Reclaim the highs and the last overhead from spring is the only thing left in the way. What actually breaks it: 238 A daily close back below it fills the gap toward 214 and turns the whole breakout into a fake. That is the invalidation. Not a red candle. A close under the pocket. Here is the part most people miss. RSI cooled from overbought to the low 60s while price only gave back a fraction of the move. Momentum reset without the trend breaking. That is the healthiest thing a chart can do after a vertical leg. The breakout was the easy part to post. The retest is the part that decides who was actually right.
0:32

Everyone who chased $TAO at 276 is now watching it in the red and wondering if the move is over. It is not. It is doing the one thing a real breakout is supposed to do. Price ran to 276, then pulled back into 252. That is not weakness. That is the retest. The level that has to hold: 252 That is the 0.382 of this whole leg, and price is sitting right on it. Hold this zone on daily closes and the pullback is just the trend catching its breath. The line under it that still keeps bulls in control: 238 The 0.618 golden pocket. As long as dips get bought between here and there, the higher low structure stays intact and the uptrend is fine. Where it goes if the retest holds: 276, then open air Reclaim the highs and the last overhead from spring is the only thing left in the way. What actually breaks it: 238 A daily close back below it fills the gap toward 214 and turns the whole breakout into a fake. That is the invalidation. Not a red candle. A close under the pocket. Here is the part most people miss. RSI cooled from overbought to the low 60s while price only gave back a fraction of the move. Momentum reset without the trend breaking. That is the healthiest thing a chart can do after a vertical leg. The breakout was the easy part to post. The retest is the part that decides who was actually right.

2xnmore

20,821 görüntüleme • 1 ay önce

Using Claude Fable 5, I built a model that predicts the entire 2026 FIFA world cup.. every single game, not just the final.. so let me break the whole thing down. what it does, how it works, and exactly how i built it.. #1 First what it does: it predicts all 104 games of the tournament. not just who lifts the trophy, but every group match, every knockout, the full path from the round of 32 to the final.. everything lands in one dashboard: > group stage, every match with each team's win % and the chance of a draw > standings, how all 12 groups are projected to finish > bracket, the full knockout tree with each team's odds of advancing > champion odds, who's most likely to actually win it all and it doesn't freeze after one prediction. the moment a real game is played, it locks that result in and re-runs everything around it. so the odds move live as the tournament goes, week by week you watch favorites rise and contenders collapse. #2. How it works: the core idea is simple. the model only ever predicts one thing, a single match. the real trick is the repetition. it learns from decades of match history, then plays the whole tournament out from the first game to the final, tens of thousands of times. each run it records who advanced and who won. do that enough and you stop getting one guess and start getting real odds, one team lifts the trophy in maybe 14% of the runs, another in 9%, and so on. #3. So, how i built it ? i didn't hand-write most of the code. i broke the project into 4 pieces, described each one to fable, and let it build while i focused on getting the football logic exactly right. - The data every international match going back over a century, around 50,000 games, plus each team's elo rating, which is the truest measure of strength, and the official 2026 schedule. garbage data means garbage predictions, so this part mattered most. - The features i turned that raw history into signals the model can learn from, the elo gap between the two teams, recent form, goals scored and conceded, and a home boost for the hosts, usa, canada and mexico. - The model for each match it predicts the expected goals for both sides, then turns that into win, draw and loss probabilities plus a likely scoreline. that's what feeds the simulation. - The tournament engine this was the hard part. the 2026 world cup is brand new, 48 teams, 12 groups, a round of 32 that's never existed before, and 8 "best third-placed" teams that slot into the bracket by a fixed fifa table. even the group tiebreakers changed this year, head to head now counts before goal difference. get any of it wrong and the whole bracket falls apart, so i built it carefully and tested the format until it was exact, then wrapped it in a simulation loop that plays the tournament out tens of thousands of times. and the last piece, the live part. as real results come in, they get locked, and only the unplayed games get re-simulated. that's what makes it a living model instead of a one-time prediction. all of it outputs to a clean dashboard you can actually read and screenshot.. right now, before kickoff, it already has a clear favorite to lift the trophy.. 👀 btw who's your pick to win the 2026 world cup?
0:20

Using Claude Fable 5, I built a model that predicts the entire 2026 FIFA world cup.. every single game, not just the final.. so let me break the whole thing down. what it does, how it works, and exactly how i built it.. #1 First what it does: it predicts all 104 games of the tournament. not just who lifts the trophy, but every group match, every knockout, the full path from the round of 32 to the final.. everything lands in one dashboard: > group stage, every match with each team's win % and the chance of a draw > standings, how all 12 groups are projected to finish > bracket, the full knockout tree with each team's odds of advancing > champion odds, who's most likely to actually win it all and it doesn't freeze after one prediction. the moment a real game is played, it locks that result in and re-runs everything around it. so the odds move live as the tournament goes, week by week you watch favorites rise and contenders collapse. #2. How it works: the core idea is simple. the model only ever predicts one thing, a single match. the real trick is the repetition. it learns from decades of match history, then plays the whole tournament out from the first game to the final, tens of thousands of times. each run it records who advanced and who won. do that enough and you stop getting one guess and start getting real odds, one team lifts the trophy in maybe 14% of the runs, another in 9%, and so on. #3. So, how i built it ? i didn't hand-write most of the code. i broke the project into 4 pieces, described each one to fable, and let it build while i focused on getting the football logic exactly right. - The data every international match going back over a century, around 50,000 games, plus each team's elo rating, which is the truest measure of strength, and the official 2026 schedule. garbage data means garbage predictions, so this part mattered most. - The features i turned that raw history into signals the model can learn from, the elo gap between the two teams, recent form, goals scored and conceded, and a home boost for the hosts, usa, canada and mexico. - The model for each match it predicts the expected goals for both sides, then turns that into win, draw and loss probabilities plus a likely scoreline. that's what feeds the simulation. - The tournament engine this was the hard part. the 2026 world cup is brand new, 48 teams, 12 groups, a round of 32 that's never existed before, and 8 "best third-placed" teams that slot into the bracket by a fixed fifa table. even the group tiebreakers changed this year, head to head now counts before goal difference. get any of it wrong and the whole bracket falls apart, so i built it carefully and tested the format until it was exact, then wrapped it in a simulation loop that plays the tournament out tens of thousands of times. and the last piece, the live part. as real results come in, they get locked, and only the unplayed games get re-simulated. that's what makes it a living model instead of a one-time prediction. all of it outputs to a clean dashboard you can actually read and screenshot.. right now, before kickoff, it already has a clear favorite to lift the trophy.. 👀 btw who's your pick to win the 2026 world cup?

Axel Bitblaze 🪓

63,796 görüntüleme • 4 ay önce

A soul tie is what happens when a physical and emotional connection creates a bond with someone outside of a holy covenant. When the relationship ends, the connection does not automatically go with it. You pray it back: call every part of yourself you gave away back into position, and release every part of them that attached to you. People confess it, pray it, and something physically breaks loose. That is not strange. That is just the soul being untangled from what was never supposed to stick.
0:55

A soul tie is what happens when a physical and emotional connection creates a bond with someone outside of a holy covenant. When the relationship ends, the connection does not automatically go with it. You pray it back: call every part of yourself you gave away back into position, and release every part of them that attached to you. People confess it, pray it, and something physically breaks loose. That is not strange. That is just the soul being untangled from what was never supposed to stick.

Lance Wallnau

19,110 görüntüleme • 4 ay önce

you numbered the steps so it could not get lost it does not get lost that list was written for a model that wandered, and this one plans better than the list does, so every number you added is a place where it is no longer allowed, to be smarter than you state the goal, the constraints, and how you will know it worked keep the numbers only where exactly one sequence is safe, which is deletes, auth and compliance the procedure felt like rigour it is a cage built to the dimensions of a model that is not in the room
1:21

you numbered the steps so it could not get lost it does not get lost that list was written for a model that wandered, and this one plans better than the list does, so every number you added is a place where it is no longer allowed, to be smarter than you state the goal, the constraints, and how you will know it worked keep the numbers only where exactly one sequence is safe, which is deletes, auth and compliance the procedure felt like rigour it is a cage built to the dimensions of a model that is not in the room

Lagosaur

14,857 görüntüleme • 2 ay önce

your agent reviewing its own work is not a check. it is a second opinion from the same source. this is the most common gap in agent systems and it hides in plain sight, because the step exists. there is a review. it just cannot do the thing you think it does. here is the mechanism. the model produced an output from a context. you then ask the same model, holding the same context, whether that output is correct. it answers fluently, because that is what it does. and the answer is drawn from the same distribution that produced the thing being judged. same weights, same window, same blind spots. if the reason the output is wrong is something the model does not know, the review does not know it either. if the reason is something the context does not contain, the review has the same context. the failure mode and the detector share a cause. > why it feels like it works because most of the time the output is fine, and the review says fine. agreement is not evidence of detection. a reviewer that says pass on everything agrees with reality most of the time too. what you actually want to measure is what happens on the cases that are wrong. that is the only place a check earns its name, and it is exactly the place where a self-review is weakest. there is research on this. Huang and colleagues at DeepMind showed at ICLR 2024 that intrinsic self-correction, revising without external grounding, does not reliably help and often makes things worse. > what to actually do move the check outside the model. a test that runs, a schema that validates, a file that exists or does not, an exit code from something you did not write. these are not smarter than the model. they are just not correlated with it, and that is the entire value. when the judgement genuinely needs a model, at minimum use a different family. same family means shared blind spots, and frontier judges measurably inflate scores for outputs that look like their own. and split the work by kind. anything objectively checkable goes to code. only the genuinely semantic calls go to a judge, and those get a rubric written as one line. a review inside the loop tells you the model is confident. a check outside it tells you whether the work is done. save this - then read the eval setup below
0:50

your agent reviewing its own work is not a check. it is a second opinion from the same source. this is the most common gap in agent systems and it hides in plain sight, because the step exists. there is a review. it just cannot do the thing you think it does. here is the mechanism. the model produced an output from a context. you then ask the same model, holding the same context, whether that output is correct. it answers fluently, because that is what it does. and the answer is drawn from the same distribution that produced the thing being judged. same weights, same window, same blind spots. if the reason the output is wrong is something the model does not know, the review does not know it either. if the reason is something the context does not contain, the review has the same context. the failure mode and the detector share a cause. > why it feels like it works because most of the time the output is fine, and the review says fine. agreement is not evidence of detection. a reviewer that says pass on everything agrees with reality most of the time too. what you actually want to measure is what happens on the cases that are wrong. that is the only place a check earns its name, and it is exactly the place where a self-review is weakest. there is research on this. Huang and colleagues at DeepMind showed at ICLR 2024 that intrinsic self-correction, revising without external grounding, does not reliably help and often makes things worse. > what to actually do move the check outside the model. a test that runs, a schema that validates, a file that exists or does not, an exit code from something you did not write. these are not smarter than the model. they are just not correlated with it, and that is the entire value. when the judgement genuinely needs a model, at minimum use a different family. same family means shared blind spots, and frontier judges measurably inflate scores for outputs that look like their own. and split the work by kind. anything objectively checkable goes to code. only the genuinely semantic calls go to a judge, and those get a rubric written as one line. a review inside the loop tells you the model is confident. a check outside it tells you whether the work is done. save this - then read the eval setup below

Hanako

14,325 görüntüleme • 3 ay önce

Memory vs. Graphs, clearly explained! memory is great, and the ceiling arrives quietly: it stores what happened. it does not store what to do about it. six runs later your file has fifty lines, and the model reloads all of them before it does anything. Graph engineering fixes this by changing what memory is: not a place things are kept, but an edge that runs backwards. you need both, and here is the sentence that resolves the whole confusion: a store keeps what happened. an edge keeps what to do about it. ↳ a store grows with every run, and every line is reloaded before the next one ↳ an edge carries one derived rule, and the rule replaces the run that produced it Prompts → Context → Harness → Loops → Graphs the transcript goes away, the constraint stays. and the constraint is smaller, because "adapters preserve keyword args exactly" is four hundred tokens shorter than the run that proved it. the same four blocks work on anything you can cut into lanes. i pointed them at token launches on Robinhood Chain, open source, nothing leaves your terminal the trick is knowing what deserves to survive. an output is not memory. "ported the utils slice, green on first pass" tells the next run nothing it can act on. the rule you derived from it does. one thing to know before you scale it. what you write down is not what comes back. ↳ the root rules file and auto memory are re-injected from disk. they come back intact, every time ↳ path-scoped rules live in message history. they get summarized away and do not return until a matching file is read again so a rule that must persist cannot be path-scoped. move it to the root and pay the always-loaded cost, or accept that it is advisory in any long session. and the one that eats whole nights: a memory file that has never had a line deleted is not memory. it is a tax on every run you will ever make, and nobody reads it back. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this, and the repo that runs it is below ↓
0:45

Memory vs. Graphs, clearly explained! memory is great, and the ceiling arrives quietly: it stores what happened. it does not store what to do about it. six runs later your file has fifty lines, and the model reloads all of them before it does anything. Graph engineering fixes this by changing what memory is: not a place things are kept, but an edge that runs backwards. you need both, and here is the sentence that resolves the whole confusion: a store keeps what happened. an edge keeps what to do about it. ↳ a store grows with every run, and every line is reloaded before the next one ↳ an edge carries one derived rule, and the rule replaces the run that produced it Prompts → Context → Harness → Loops → Graphs the transcript goes away, the constraint stays. and the constraint is smaller, because "adapters preserve keyword args exactly" is four hundred tokens shorter than the run that proved it. the same four blocks work on anything you can cut into lanes. i pointed them at token launches on Robinhood Chain, open source, nothing leaves your terminal the trick is knowing what deserves to survive. an output is not memory. "ported the utils slice, green on first pass" tells the next run nothing it can act on. the rule you derived from it does. one thing to know before you scale it. what you write down is not what comes back. ↳ the root rules file and auto memory are re-injected from disk. they come back intact, every time ↳ path-scoped rules live in message history. they get summarized away and do not return until a matching file is read again so a rule that must persist cannot be path-scoped. move it to the root and pay the always-loaded cost, or accept that it is advisory in any long session. and the one that eats whole nights: a memory file that has never had a line deleted is not memory. it is a tax on every run you will ever make, and nobody reads it back. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this, and the repo that runs it is below ↓

Hanako

48,225 görüntüleme • 2 ay önce

SOMEONE VIBE CODED AN AI THAT REDESIGNS YOUR ROOM USING FURNITURE YOU CAN ACTUALLY GO BUY every other ai interior tool gives you a beautiful render of furniture that doesnt exist you screenshot it, you love it, and then you cant buy a single thing in it this one matches every piece to a real product > upload a photo of your room and pick a style > it redesigns the space and swaps in furniture that actually exists > every item in the render links to somewhere you can order and buy it > the sidebar builds you a full shopping list with the total cost > each piece is priced individually and tagged with the store its from > every item has 3 alternate options, so if you dont like the nightstand you swap it and the render updates > theres a before and after compare toggle, and you can type plain english to change anything in the room > you set your country so it only shows you stores that actually ship to you nobody knows what furniture actually looks like in their own room until its already delivered, and this fixes that
0:41

SOMEONE VIBE CODED AN AI THAT REDESIGNS YOUR ROOM USING FURNITURE YOU CAN ACTUALLY GO BUY every other ai interior tool gives you a beautiful render of furniture that doesnt exist you screenshot it, you love it, and then you cant buy a single thing in it this one matches every piece to a real product > upload a photo of your room and pick a style > it redesigns the space and swaps in furniture that actually exists > every item in the render links to somewhere you can order and buy it > the sidebar builds you a full shopping list with the total cost > each piece is priced individually and tagged with the store its from > every item has 3 alternate options, so if you dont like the nightstand you swap it and the render updates > theres a before and after compare toggle, and you can type plain english to change anything in the room > you set your country so it only shows you stores that actually ship to you nobody knows what furniture actually looks like in their own room until its already delivered, and this fixes that

Om Patel

13,428 görüntüleme • 3 ay önce

this video is the CLEAREST explanation of how claude skills + AI agents work and how to use them most people set up an AI agent and wonder why it keeps disappointing them. the context window is everything context is what the model assembles before it takes any action. think of it like everything the agent needs to read before it does anything. the quality of what goes in determines the quality of what comes out. the models are genuinely really good right now. claude and gpt are exceptional. the variable is almost always the context you give them. 1. agent.md files are mostly unnecessary every single line you put in an agent.md file gets added to every single conversation you have with your agent. a 1000 line file is around 7000 tokens burning on every run. the model already knows to use react. it can read your codebase. save the agent.md for proprietary information specific to your company that the model genuinely cannot know on its own. 2. skills are the actual unlock a skill.md file works differently. what loads into context is only the name and description, around 50 tokens. the full instructions only appear when the agent recognizes it needs that skill. so instead of 7000 tokens on every run you have 50. and the agent stays sharp because the context window stays lean. the closer you get to filling the context window the worse the agent performs, same way you perform worse when someone dumps 10 things on you at once. 3. here is how to actually build a skill the right way most people identify a workflow and immediately try to write the skill. what you want to do instead is run the workflow by hand with the agent first. walk it through every single step. tell it what to check, what good looks like, what bad looks like. correct it in real time. once you have had a full successful run from start to finish, tell the agent to review everything it just did and write the skill itself. it writes a better skill than you will because it has the full context of what actually worked in practice not in theory. 4. recursively building skills is how you go from frustrated to reliable when the skill breaks, and it will break, ask the agent exactly why it failed. it will tell you specifically what went wrong. fix it together in that same conversation. then tell it to update the skill file so that failure mode never happens again. ross mike did this five times with his youtube report generator. it now pulls from eight different data sources and runs flawlessly every single time without him touching it. 5. sub agents are something you earn not something you set up on day one start with one agent. build one workflow. turn it into one skill. once that works add another. ross mike has five sub agents now covering marketing, business, personal and more. it took months to get there and every single one exists because a workflow proved it deserved to exist. the people who set up 15 sub agents on day one and wonder why nothing works skipped all the steps that make the thing actually run. 6. your workflow is the thing the model cannot get anywhere else the model has been trained on everything. it knows more than you about most things. what it does not have is your specific process, your taste, your way of doing things. that is what skills capture. that is what makes your agent actually useful versus a generic one. downloading someone else's skill means downloading their context onto your setup and it will not work the way you want it to because it was never built around how you work. this is the clearest explanation of how agents actually work i have heard. Micky runs this stuff every single day and the results show it. full episode is now live on The Startup Ideas Podcast (SIP) 🧃 where you get your pods people charge for this sorta stuff i give away the sauce for free i just want you to win watch
35:25

this video is the CLEAREST explanation of how claude skills + AI agents work and how to use them most people set up an AI agent and wonder why it keeps disappointing them. the context window is everything context is what the model assembles before it takes any action. think of it like everything the agent needs to read before it does anything. the quality of what goes in determines the quality of what comes out. the models are genuinely really good right now. claude and gpt are exceptional. the variable is almost always the context you give them. 1. agent.md files are mostly unnecessary every single line you put in an agent.md file gets added to every single conversation you have with your agent. a 1000 line file is around 7000 tokens burning on every run. the model already knows to use react. it can read your codebase. save the agent.md for proprietary information specific to your company that the model genuinely cannot know on its own. 2. skills are the actual unlock a skill.md file works differently. what loads into context is only the name and description, around 50 tokens. the full instructions only appear when the agent recognizes it needs that skill. so instead of 7000 tokens on every run you have 50. and the agent stays sharp because the context window stays lean. the closer you get to filling the context window the worse the agent performs, same way you perform worse when someone dumps 10 things on you at once. 3. here is how to actually build a skill the right way most people identify a workflow and immediately try to write the skill. what you want to do instead is run the workflow by hand with the agent first. walk it through every single step. tell it what to check, what good looks like, what bad looks like. correct it in real time. once you have had a full successful run from start to finish, tell the agent to review everything it just did and write the skill itself. it writes a better skill than you will because it has the full context of what actually worked in practice not in theory. 4. recursively building skills is how you go from frustrated to reliable when the skill breaks, and it will break, ask the agent exactly why it failed. it will tell you specifically what went wrong. fix it together in that same conversation. then tell it to update the skill file so that failure mode never happens again. ross mike did this five times with his youtube report generator. it now pulls from eight different data sources and runs flawlessly every single time without him touching it. 5. sub agents are something you earn not something you set up on day one start with one agent. build one workflow. turn it into one skill. once that works add another. ross mike has five sub agents now covering marketing, business, personal and more. it took months to get there and every single one exists because a workflow proved it deserved to exist. the people who set up 15 sub agents on day one and wonder why nothing works skipped all the steps that make the thing actually run. 6. your workflow is the thing the model cannot get anywhere else the model has been trained on everything. it knows more than you about most things. what it does not have is your specific process, your taste, your way of doing things. that is what skills capture. that is what makes your agent actually useful versus a generic one. downloading someone else's skill means downloading their context onto your setup and it will not work the way you want it to because it was never built around how you work. this is the clearest explanation of how agents actually work i have heard. Micky runs this stuff every single day and the results show it. full episode is now live on The Startup Ideas Podcast (SIP) 🧃 where you get your pods people charge for this sorta stuff i give away the sauce for free i just want you to win watch

GREG ISENBERG

195,059 görüntüleme • 7 ay önce

AI in CAD just went from “make me a part” to “build me the damn machine.” Using Autodesk Fusion MCP, the agent is working inside a real mechanical assembly with a chassis, shafts, gears, bearings, brackets, repeated hardware, and multiple components that all have to stay aligned with each other. That is a completely different problem from generating one clean 3D object. You can inspect the assembly from different angles, isolate sections, dig into individual joints, and keep working on the same engineering project instead of starting from another prompt-generated blob. The important part is not that AI can create geometry anymore. It is that it can operate inside the structure of a real CAD project where every component depends on what is around it. Once this gets reliable, a lot of mechanical CAD starts looking suspiciously simple: describe the system, inspect what the agent built, then tell it what to change.
0:29

AI in CAD just went from “make me a part” to “build me the damn machine.” Using Autodesk Fusion MCP, the agent is working inside a real mechanical assembly with a chassis, shafts, gears, bearings, brackets, repeated hardware, and multiple components that all have to stay aligned with each other. That is a completely different problem from generating one clean 3D object. You can inspect the assembly from different angles, isolate sections, dig into individual joints, and keep working on the same engineering project instead of starting from another prompt-generated blob. The important part is not that AI can create geometry anymore. It is that it can operate inside the structure of a real CAD project where every component depends on what is around it. Once this gets reliable, a lot of mechanical CAD starts looking suspiciously simple: describe the system, inspect what the agent built, then tell it what to change.

Rina

14,154 görüntüleme • 2 ay önce

Nobody is paying for the stone. They are paying for the second you can't see inside it. And that second is now the most expensive thing on your feed. Read this very slowly, because it will change how you scroll. Somewhere right now a rock is being split open, and a few million people are about to feel something that has no name in any language. A jolt. A hot little spike of greed and wonder. The oldest button in the human brain, pressed on purpose. Here is the part nobody tells you. That button has been industrialized. There is a whole machine behind that feeling. Geologists, jewelers, shell companies, 3D artists, render farms, editors with stopwatches, and now models that can paint a rainbow into the inside of any object on Earth in about forty seconds. The machine does not care whether the thing inside is real. It cares whether you hold your breath. So let me ask the only question that matters. If you can't tell a miracle from a render, what exactly are you watching for? Stay with me. This is going to get uncomfortable, then useful, then genuinely beautiful. In that order. PART 1. THE PHYSICS THAT SHOULD NOT WORK Start with the real thing, because the real thing is stranger than any fake. Precious opal is not a crystal. That sentence alone should bother you. Almost every gemstone you know is a crystal, an orderly lattice of atoms repeating in every direction, like soldiers on a parade ground. Opal skips the parade entirely. Opal is made of microscopic spheres of silica. Tiny glass beads, each one a few hundred nanometers across. Stacked together like oranges in a crate, in layers, in a pattern so regular that it works as a natural diffraction grating. Read that again. Nature stacked glass marbles smaller than a wavelength of light, and by accident built an optical instrument. When white light slips between those spheres, it bends and interferes with itself. Some wavelengths cancel. Others reinforce. Which color survives depends on the size of the spheres and the angle you tilt your head. Larger spheres push the answer toward red. Smaller ones toward violet and blue. There is no pigment in the good stuff. No dye. No coloring agent at all. The color is structure. Geometry doing the job that chemistry usually does. Butterfly wings do the same trick. Peacock feathers too. Soap bubbles. The deep blue of some beetles. Evolution found structural color a few hundred million years before we had a word for physics, and a patch of sinking silica gel found it too. Now think about how that structure forms, because this is where the story turns. Silica-rich water seeps down through ancient rock. Slowly. Slower than you can picture. It settles into cracks, voids, the hollow spaces left by something that used to live there. Then it waits. Over spans of time that make human history look like a hiccup, the silica spheres sort themselves by size and settle into layers. Sort. By size. On their own. In the dark. Underground. If the spheres come out uneven, you get milky, dull, common opal. Pretty. Cheap. Nobody writes songs about it. If they come out near perfect, uniform to within a few percent, stacked in clean layers, you get play of color. The full spectrum rolling across the surface like a living thing. The gap between those two outcomes is a matter of how evenly a few billion invisible beads settled. And that tiny difference in order is the difference between a few dollars a carat and a number that sounds fake when you say it out loud. Black opal from Lightning Ridge in Australia can sell for tens of thousands of dollars per carat at the top end. Exceptional specimens go far beyond that. A stone the size of a thumbnail can cost more than a car. All of it from sand and water and patience. And here is the cruel twist the whole industry runs on. You cannot know what is inside until it is opened. PART 2. THE GAMBLE THAT BUILT AN ECONOMY Imagine you are a miner. You have a lump of ordinary looking rock. Brown. Dull. It looks like every other lump of ordinary looking rock on the ground. Inside it is either nothing, or a fortune, and no instrument you own can tell you which. So what do you do? You make a tiny cut. A window. A sliver of surface ground away to peek at the layer beneath. Or you trust the shape, the weight, the way the host rock behaves, the way your grandfather taught you to read it. Or you just gamble and break it. This is not a metaphor. This is how entire mining towns work. Whole families have lived and died on the thin line between a worthless nodule and a once in a lifetime find. Economists have a clean name for this. Information asymmetry. The seller knows something the buyer doesn't, or nobody knows anything, and the price becomes a bet on the unknown. Psychologists have a different name for it. Variable reward. It is the same mechanism that makes slot machines the most profitable real estate in any casino. The same mechanism that makes you pull down to refresh your feed. Not knowing what comes next is more intoxicating than knowing it will be good. Your brain releases more dopamine anticipating an uncertain reward than receiving a guaranteed one. Let that sink in. The wait is the drug. The reveal is just the receipt. Now put those two facts side by side. One. Nature hides its most valuable color behind plain rock, and the only way to find it is to open the rock. Two. The human brain is addicted to exactly that moment of opening. Somebody was always going to notice that these two things fit together like a key and a lock. PART 3. THE PEOPLE WHO NOTICED There is a genre now. You know it even if you never named it. A person, a rough object, a single tool, and a pause that stretches just long enough to make your thumb stop moving. Then the reveal. Then the next one. Billions of views, collected one held breath at a time. Creators figured out something the diamond cartel took a century to learn. Scarcity is a story, and a story can be filmed. The format is brutally simple, which is exactly why it works. Beat one: ordinary. The most boring object you can find. The boredom is not a flaw, it is the setup. You cannot feel a surprise unless you first agreed to expect nothing. Beat two: effort. A tool. A strike. Something that costs the creator a little sweat and gives the viewer a little dread. Will it crack. Will it ruin the thing. Tension needs a stake. Beat three: the opening. Not a slow fade. A cut. A change of state. From dull to impossible in a single frame. Beat four: the color. Always the color. Because color is the one thing your visual cortex cannot ignore. Saturated, high contrast, glowing from within. Your eyes treat it as a signal. Fruit. Fire. Water. Poison. Something that matters. Then the cycle repeats and the algorithm does the rest. This is not manipulation in the cheap sense. It is something older and more honest. It is the structure of every treasure hunt told around a fire since language began. We just compressed it to forty seconds and put a recommendation engine behind it. And here is where it gets interesting. Because compressed, optimized, engineered attention does not stay innocent for long. PART 4. THE FIRST LAW OF THE ATTENTION ECONOMY If something is valuable to watch, someone will manufacture more of it than nature can supply. That is not cynicism. It is arithmetic. Real discoveries of spectacular stones happen rarely. A top quality piece of play of color opal might surface a handful of times in a miner's career, if they are lucky. A feed, however, needs a fresh one every day. Every hour. Every few seconds, if you scroll fast. Demand is infinite. Supply is geological. You can feel how this ends. Somewhere in that gap, a different kind of supply appeared. Not mined. Made. And I mean made in more than one way, which is where this story forks into two very different roads. PART 5. ROAD ONE: THE LAB Let's take the honest fork first, because it is genuinely astonishing. In the 1960s, a French researcher named Pierre Gilson did something that sounds impossible. He grew opal. Not a look alike. Not a coating. A structure of silica spheres, ordered the way nature orders them, cultivated in a controlled environment over a long and fussy process. Think about what that means. He recreated the geometry that takes the earth an unimaginable span of time to produce, and he did it on a schedule. Today, lab grown opal can show play of color that rivals natural stone. It is chemically the same family of material. It is built from the same idea. And a trained eye with the right instruments can still tell it apart, because the order of a lab structure often looks too regular, too tidy, with a pattern that reads like a snakeskin or a lizard's scales when you look closely. Nature is messy. The lab is neat. That is often the tell. Is lab opal fake? That depends entirely on what you were told when you bought it. If the label says created in a lab, it is a legitimate product with a legitimate price. If the label says nothing, or says the opposite, it is fraud. The material is identical. The honesty is the whole difference. Hold that thought. Honesty will be the theme of everything that follows. PART 6. ROAD TWO: THE PIXELS Now the second fork. The one that should make you sit up. There is a cheaper way to manufacture a rainbow than growing silica spheres over months. You simply don't have the stone at all. A render is a stone with no weight. A synthetic reveal needs no mine, no hammer, no riverbed, no luck. It needs a prompt, a patient artist, and a machine that has studied millions of real photographs of real opals and learned exactly what your eyes expect them to look like. And here is the unsettling part. The machine has gotten better at the look of opal than most humans are at recognizing it. Why? Because play of color is hard for a camera to capture honestly, but easy for a model to exaggerate beautifully. A photograph of real opal is constrained by lighting, by angle, by physics. A generated image is constrained only by what looks good. And what looks good, statistically, is more saturated, more symmetrical, more luminous, more perfect than the real thing. Read that twice. The fake can be more convincing than the original precisely because it is more beautiful than the original. That is a strange new problem. Humans spent all of history learning to distrust the ugly, the flawed, the suspicious. Now the danger is the flawless. When something looks too good, your instinct says amazing. It should say wait. PART 7. THE TELLS THAT STILL WORK Good news. You are not helpless. Real and rendered things leave different fingerprints, and once you learn them, you can't unsee them. Here is the toolkit. Save it. One. Check the physics of the color. Real play of color changes with angle. Tilt the stone and the pattern moves, flashes, goes dark, and returns somewhere else. That is the diffraction grating at work. A genuine specimen under a moving hand should shimmer differently from frame to frame, with flashes that appear and vanish as the light direction shifts. A fake often behaves like a sticker. The pattern stays glued to the surface while the viewpoint changes. Or it changes in ways that don't match the lighting around it. Two. Check the boundary. Where the colored zone meets the host rock matters enormously. In nature, that edge is irregular, grainy, full of transition, geological in its messiness. Look for places where color bleeds into stone in ways that look earned. In a render, boundaries tend to be suspiciously smooth, or suspiciously dramatic, or oddly clean, like a painting that stopped at the frame. Three. Check the scale of the detail. Real stones contain structure at every magnification. Cracks inside cracks. Inclusions. Tiny flaws no artist would think to add. Zoom in and the complexity keeps going. Generated imagery often runs out of detail. Past a certain zoom, it melts into smooth gradient soup. Four. Check the hands and the objects touching the stone. Contact is the hardest thing to fake. Weight, shadow, compression of skin, the way fingers curve around real mass. When a hand grips something heavy, the whole arm knows it. When a hand grips something that isn't there, there is always a tiny lie in the posture. Five. Check the audio. A real strike on real stone has a specific sound. Dense, uneven, with a tail that rings differently depending on what is inside. Foley artists can fake it, but they rarely fake all of it, all the time, perfectly synced to every impact. Six, and this is the one that matters most. Check the source. Who posted it. When. Where else does it appear. Is there a chain of custody, a mine, a seller, an auction record, a name you can look up. A miracle with no provenance is not a miracle. It is content. PART 8. WHY EVEN THE FAKE IS DOING SOMETHING REAL Here is where I'm going to disagree with the people who want to burn all of it down. Because a rendered gem is not worthless. It is doing real work on your nervous system, and that work deserves to be understood rather than sneered at. Humans have always built illusions to reach feelings that reality rations out too slowly. Theater. Cathedrals with stained glass. Fireworks. Cinema. Nobody calls a movie a lie because the spaceship wasn't real. The deal is clear. You know it is made. You consent to the feeling. The problem is never the illusion. The problem is the missing label. A simulation announced as a simulation is art. A simulation disguised as discovery is a weapon. Same pixels. Opposite ethics. The only thing that changes is whether you were told the truth before you felt the thing. So when a clip shows up carrying a title that admits what it is, that is not a trick. That is the rarest and most underrated move in modern media. Someone said the quiet part first. And the feeling you get afterward is cleaner for it, because you are not spending the next hour wondering if you were played. Remember that. In a feed full of costumes, the one wearing a name tag is the one worth trusting. PART 9. THE MONEY NOBODY MENTIONS Let's follow the cash, because it explains almost everything. A gemstone video earns attention. Attention earns reach. Reach earns three kinds of income. Direct. Ad revenue, subscriptions, sponsorships. The obvious pipe. Indirect. Traffic to a shop. A storefront selling polished stones, jewelry, kits, tools, mystery boxes. The video is a free sample. The checkout is the business. And third, the quiet one. Authority. Once you have a few million views, you are an expert, whether or not you ever held a real specimen. Authority converts into everything else. Now here is the part that should make your stomach tighten. Mystery boxes. Grab bags. Dig it yourself kits. Sealed stones sold with the promise that something spectacular might hide inside. Think about what that product is. It is the variable reward machine from earlier, sold back to you as a physical object, priced above its contents, with a promotional video showing someone else winning. The video shows the jackpot. The box contains the expected value. And the expected value of a cheap sealed rock is almost always a cheap rock. This is not an accusation against every seller. Plenty of honest dealers sell real stones with real descriptions at fair prices, and the best of them will happily tell you exactly what the odds are. But the structure rewards the opposite behavior, which is why the advice is simple. If the odds are hidden, assume they are bad. If the reveal is the product, assume the reveal is staged. If the price is a fraction of what the footage implies, assume the footage lies. PART 10. WHAT AN EXPERT ACTUALLY DOES Want to know how a real gemologist handles a mystery stone? Slowly. Boringly. With tools that make for terrible content. First, a loupe. Ten power magnification. They look for the structure of the surface, the pattern of the color, the telltale features that separate natural from lab. Second, a scale and a water test, to measure specific gravity. Real opal has a density in a known range. Plastic, glass, and some imitations land somewhere else entirely. Third, behavior under light. Ultraviolet. Polarized. Different materials answer these differently. A stone has a personality, and light is how you interview it. Fourth, the paper trail. Where was it found. Who sold it. What documents exist. In a serious market, a stone without a story is priced like a stone without a pedigree. Notice something. None of this is dramatic. None of it fits in forty seconds. None of it ends with a gasp. That is the entire problem and the entire lesson. Truth is slow. Attention is fast. The two are in a permanent tug of war, and whoever controls the clock controls the room. PART 11. THE CHEAPEST SUPERPOWER Let me give you the one habit that protects you from nearly everything in this post. When a clip makes you feel a surge, wait three seconds before you share it. That is it. Three seconds. Not to kill the feeling. Just to let the thinking part of your brain catch up with the feeling part. Ask two questions. What exactly am I being asked to believe? Who benefits if I believe it? If the answer to the first is something extraordinary, and the answer to the second is a person with a link in their bio, slow down. This is not paranoia. It is hygiene. The same way you wash your hands, not because every surface is deadly, but because the cost of doing it is nothing and the cost of not doing it is occasionally everything. You can still enjoy the show. You can still gasp. You just stop paying for it with your trust. PART 12. THE DEEPER THING Zoom out for a moment. We are living through the first era in history where seeing is no longer believing, and nobody has updated the instruction manual. For nearly all of human existence, an image was evidence. A photograph meant a thing happened. A video meant it happened in front of a lens. That rule held for roughly a hundred and fifty years, and entire legal systems, newsrooms, and personal relationships were built on top of it. That rule just expired. It did not expire with a bang. It expired the way a milk carton does. Quietly, in the dark, while everyone kept drinking. What replaces it? Not paranoia. A society that disbelieves everything is as broken as one that believes everything. Cynics are easy to fool too. They just get fooled in the other direction, dismissing real things because real things can now be dismissed. What replaces it is provenance. Receipts. Chains of custody. Systems that say who made this, with what, when, and who vouches for it. Cryptographic signatures at the moment of capture. Labels that survive a screenshot. Platforms that tell you plainly what is synthetic. We are building that. Slowly. Unevenly. Not fast enough. Until then, the responsibility falls where it always has, on the one person standing between a clip and a share button. You. PART 13. THE BEAUTIFUL PART I PROMISED Now the good news. I said this would turn beautiful, and I keep my promises. Strip away the money, the algorithm, the renders, the grifters. Underneath all of it sits a thing that is still true, still real, still waiting for you. The universe really does hide color inside ordinary things. It is not a trick. It is not marketing. It is the actual shape of reality. Plain gray granite carries minerals that glow under ultraviolet light like a hidden city. A dull pebble on a riverbank can be a fragment of a thing that took a hundred million years to form. A drop of water on a leaf bends light into a full spectrum, for free, every morning, to nobody in particular. The only reason a fake gemstone can move you is that a real one once moved somebody first. The counterfeit borrows its power from the original. Every render of a rainbow is a love letter to a rainbow that existed before the first camera. So here is the invitation. Go find the real thing. Not on a feed. In a museum with a velvet tray and a rock that is allowed to be dull. At a gem and mineral show where a grizzled dealer will hand you a lump and tell you what it is in four honest sentences. On a riverbank, with your own eyes, picking up stones and turning them in the sun. Hold a real opal under a lamp and rotate your wrist. Watch the color fall in and out of existence. Feel the small private shock of it. That shock cannot be rendered, because it is not in the pixels. It is in the fact that you are standing in the same universe as the thing, breathing the same air, aware that it took longer to make than your entire species has existed. That feeling is what the machine is trying to imitate. It is also what the machine can never deliver. PART 14. THE RULES, CONDENSED Screenshot this. Send it to the person in your family who forwards everything. One. Beauty is not evidence. The more perfect it looks, the more carefully you should look. Two. Color that doesn't move with the light is color that was never in the room. Three. A miracle without a source is just entertainment. Enjoy it as that. Four. If the odds are hidden, the odds are bad. Five. Labels are a form of respect. Trust the people who use them. Six. Wait three seconds. Ask who benefits. Seven. Support the slow experts. Miners, gemologists, dealers, museum curators, the people whose work doesn't fit in a clip. They are the reason any of this is real. Eight. Let yourself feel wonder anyway. Cynicism is not intelligence. Wonder with open eyes is the highest form of it. PART 15. WHAT HAPPENS NEXT Let me make a prediction, and you can come back and check me. Within a few years, the question will not be whether a clip like this is real. The question will be whether it is labeled. Real footage will carry cryptographic proof of origin, and its value will rise because of it. Synthetic footage will carry honest labels, and its value will hold because of that, as art, as entertainment, as a legitimate craft with real skill behind it. The dangerous category will be the middle. The unlabeled. The ambiguous. The thing that wants to be believed without earning it. Creators who label honestly will build the longest careers. Platforms that enforce labels will keep their users. Audiences who learn the tells will become impossible to exploit, and the exploiters will quietly move on to easier targets. The slow, truthful, boring work of verification will turn out to be the most valuable skill of the decade. You can start building it today, one three second pause at a time. PART 16. A SMALL EXPERIMENT FOR TONIGHT Try this before you sleep. Open the next clip that gives you that jolt. Before the reveal lands, say out loud what you expect to see. Then watch it. Then ask yourself what you actually felt, and where in your body. Most people discover the feeling arrives before the information does. The chest tightens a half second ahead of the facts. That gap is the whole game, and now that you know it exists, nobody can use it on you without your permission. Do it for a week. You will start noticing patterns everywhere, in headlines, in product launches, in the way a stranger tells a story. The trick is always the same. Build a gap. Fill it with feeling. Sell the feeling. You were never the audience. You were the product. Now you are the editor. THE LAST THING Let me bring this home. We began with a rock and a second you can't see inside it. Here is what that second really is. It is the gap between what you expect and what you get. It is the entire engine of curiosity. It is the reason anyone ever picked up a stone, cracked it, and looked. It is the oldest question our species knows how to ask: what is hiding in there? A civilization is the sum of everyone who refused to stop asking. The technology that fakes the answer has not killed the question. It has only made the question more valuable. Because now, when you finally do find something real, something unlabeled by any algorithm, unpolished by any editor, unrendered, unsponsored, and entirely indifferent to whether you are watching, it will mean more than it ever did. Real wonder is going to become scarce. And scarce things are the ones worth protecting. So protect it. Teach it. Share the real things generously and the fake things honestly. Keep your eyes open and your heart open and your thumb a little slower than the feed wants it to be. The stone is just the container. The color was always yours. The best feeds of the next decade will not be the ones that fool you best. They will be the ones that let you stay amazed without ever feeling cheated. Build for that. Follow that. Reward that. Everything else is just noise. If this changed how you'll watch the next one, repost the first line, and tell me one thing in the replies. What is the most beautiful real thing you have ever held in your hands? I'll read every answer.
1:52

Nobody is paying for the stone. They are paying for the second you can't see inside it. And that second is now the most expensive thing on your feed. Read this very slowly, because it will change how you scroll. Somewhere right now a rock is being split open, and a few million people are about to feel something that has no name in any language. A jolt. A hot little spike of greed and wonder. The oldest button in the human brain, pressed on purpose. Here is the part nobody tells you. That button has been industrialized. There is a whole machine behind that feeling. Geologists, jewelers, shell companies, 3D artists, render farms, editors with stopwatches, and now models that can paint a rainbow into the inside of any object on Earth in about forty seconds. The machine does not care whether the thing inside is real. It cares whether you hold your breath. So let me ask the only question that matters. If you can't tell a miracle from a render, what exactly are you watching for? Stay with me. This is going to get uncomfortable, then useful, then genuinely beautiful. In that order. PART 1. THE PHYSICS THAT SHOULD NOT WORK Start with the real thing, because the real thing is stranger than any fake. Precious opal is not a crystal. That sentence alone should bother you. Almost every gemstone you know is a crystal, an orderly lattice of atoms repeating in every direction, like soldiers on a parade ground. Opal skips the parade entirely. Opal is made of microscopic spheres of silica. Tiny glass beads, each one a few hundred nanometers across. Stacked together like oranges in a crate, in layers, in a pattern so regular that it works as a natural diffraction grating. Read that again. Nature stacked glass marbles smaller than a wavelength of light, and by accident built an optical instrument. When white light slips between those spheres, it bends and interferes with itself. Some wavelengths cancel. Others reinforce. Which color survives depends on the size of the spheres and the angle you tilt your head. Larger spheres push the answer toward red. Smaller ones toward violet and blue. There is no pigment in the good stuff. No dye. No coloring agent at all. The color is structure. Geometry doing the job that chemistry usually does. Butterfly wings do the same trick. Peacock feathers too. Soap bubbles. The deep blue of some beetles. Evolution found structural color a few hundred million years before we had a word for physics, and a patch of sinking silica gel found it too. Now think about how that structure forms, because this is where the story turns. Silica-rich water seeps down through ancient rock. Slowly. Slower than you can picture. It settles into cracks, voids, the hollow spaces left by something that used to live there. Then it waits. Over spans of time that make human history look like a hiccup, the silica spheres sort themselves by size and settle into layers. Sort. By size. On their own. In the dark. Underground. If the spheres come out uneven, you get milky, dull, common opal. Pretty. Cheap. Nobody writes songs about it. If they come out near perfect, uniform to within a few percent, stacked in clean layers, you get play of color. The full spectrum rolling across the surface like a living thing. The gap between those two outcomes is a matter of how evenly a few billion invisible beads settled. And that tiny difference in order is the difference between a few dollars a carat and a number that sounds fake when you say it out loud. Black opal from Lightning Ridge in Australia can sell for tens of thousands of dollars per carat at the top end. Exceptional specimens go far beyond that. A stone the size of a thumbnail can cost more than a car. All of it from sand and water and patience. And here is the cruel twist the whole industry runs on. You cannot know what is inside until it is opened. PART 2. THE GAMBLE THAT BUILT AN ECONOMY Imagine you are a miner. You have a lump of ordinary looking rock. Brown. Dull. It looks like every other lump of ordinary looking rock on the ground. Inside it is either nothing, or a fortune, and no instrument you own can tell you which. So what do you do? You make a tiny cut. A window. A sliver of surface ground away to peek at the layer beneath. Or you trust the shape, the weight, the way the host rock behaves, the way your grandfather taught you to read it. Or you just gamble and break it. This is not a metaphor. This is how entire mining towns work. Whole families have lived and died on the thin line between a worthless nodule and a once in a lifetime find. Economists have a clean name for this. Information asymmetry. The seller knows something the buyer doesn't, or nobody knows anything, and the price becomes a bet on the unknown. Psychologists have a different name for it. Variable reward. It is the same mechanism that makes slot machines the most profitable real estate in any casino. The same mechanism that makes you pull down to refresh your feed. Not knowing what comes next is more intoxicating than knowing it will be good. Your brain releases more dopamine anticipating an uncertain reward than receiving a guaranteed one. Let that sink in. The wait is the drug. The reveal is just the receipt. Now put those two facts side by side. One. Nature hides its most valuable color behind plain rock, and the only way to find it is to open the rock. Two. The human brain is addicted to exactly that moment of opening. Somebody was always going to notice that these two things fit together like a key and a lock. PART 3. THE PEOPLE WHO NOTICED There is a genre now. You know it even if you never named it. A person, a rough object, a single tool, and a pause that stretches just long enough to make your thumb stop moving. Then the reveal. Then the next one. Billions of views, collected one held breath at a time. Creators figured out something the diamond cartel took a century to learn. Scarcity is a story, and a story can be filmed. The format is brutally simple, which is exactly why it works. Beat one: ordinary. The most boring object you can find. The boredom is not a flaw, it is the setup. You cannot feel a surprise unless you first agreed to expect nothing. Beat two: effort. A tool. A strike. Something that costs the creator a little sweat and gives the viewer a little dread. Will it crack. Will it ruin the thing. Tension needs a stake. Beat three: the opening. Not a slow fade. A cut. A change of state. From dull to impossible in a single frame. Beat four: the color. Always the color. Because color is the one thing your visual cortex cannot ignore. Saturated, high contrast, glowing from within. Your eyes treat it as a signal. Fruit. Fire. Water. Poison. Something that matters. Then the cycle repeats and the algorithm does the rest. This is not manipulation in the cheap sense. It is something older and more honest. It is the structure of every treasure hunt told around a fire since language began. We just compressed it to forty seconds and put a recommendation engine behind it. And here is where it gets interesting. Because compressed, optimized, engineered attention does not stay innocent for long. PART 4. THE FIRST LAW OF THE ATTENTION ECONOMY If something is valuable to watch, someone will manufacture more of it than nature can supply. That is not cynicism. It is arithmetic. Real discoveries of spectacular stones happen rarely. A top quality piece of play of color opal might surface a handful of times in a miner's career, if they are lucky. A feed, however, needs a fresh one every day. Every hour. Every few seconds, if you scroll fast. Demand is infinite. Supply is geological. You can feel how this ends. Somewhere in that gap, a different kind of supply appeared. Not mined. Made. And I mean made in more than one way, which is where this story forks into two very different roads. PART 5. ROAD ONE: THE LAB Let's take the honest fork first, because it is genuinely astonishing. In the 1960s, a French researcher named Pierre Gilson did something that sounds impossible. He grew opal. Not a look alike. Not a coating. A structure of silica spheres, ordered the way nature orders them, cultivated in a controlled environment over a long and fussy process. Think about what that means. He recreated the geometry that takes the earth an unimaginable span of time to produce, and he did it on a schedule. Today, lab grown opal can show play of color that rivals natural stone. It is chemically the same family of material. It is built from the same idea. And a trained eye with the right instruments can still tell it apart, because the order of a lab structure often looks too regular, too tidy, with a pattern that reads like a snakeskin or a lizard's scales when you look closely. Nature is messy. The lab is neat. That is often the tell. Is lab opal fake? That depends entirely on what you were told when you bought it. If the label says created in a lab, it is a legitimate product with a legitimate price. If the label says nothing, or says the opposite, it is fraud. The material is identical. The honesty is the whole difference. Hold that thought. Honesty will be the theme of everything that follows. PART 6. ROAD TWO: THE PIXELS Now the second fork. The one that should make you sit up. There is a cheaper way to manufacture a rainbow than growing silica spheres over months. You simply don't have the stone at all. A render is a stone with no weight. A synthetic reveal needs no mine, no hammer, no riverbed, no luck. It needs a prompt, a patient artist, and a machine that has studied millions of real photographs of real opals and learned exactly what your eyes expect them to look like. And here is the unsettling part. The machine has gotten better at the look of opal than most humans are at recognizing it. Why? Because play of color is hard for a camera to capture honestly, but easy for a model to exaggerate beautifully. A photograph of real opal is constrained by lighting, by angle, by physics. A generated image is constrained only by what looks good. And what looks good, statistically, is more saturated, more symmetrical, more luminous, more perfect than the real thing. Read that twice. The fake can be more convincing than the original precisely because it is more beautiful than the original. That is a strange new problem. Humans spent all of history learning to distrust the ugly, the flawed, the suspicious. Now the danger is the flawless. When something looks too good, your instinct says amazing. It should say wait. PART 7. THE TELLS THAT STILL WORK Good news. You are not helpless. Real and rendered things leave different fingerprints, and once you learn them, you can't unsee them. Here is the toolkit. Save it. One. Check the physics of the color. Real play of color changes with angle. Tilt the stone and the pattern moves, flashes, goes dark, and returns somewhere else. That is the diffraction grating at work. A genuine specimen under a moving hand should shimmer differently from frame to frame, with flashes that appear and vanish as the light direction shifts. A fake often behaves like a sticker. The pattern stays glued to the surface while the viewpoint changes. Or it changes in ways that don't match the lighting around it. Two. Check the boundary. Where the colored zone meets the host rock matters enormously. In nature, that edge is irregular, grainy, full of transition, geological in its messiness. Look for places where color bleeds into stone in ways that look earned. In a render, boundaries tend to be suspiciously smooth, or suspiciously dramatic, or oddly clean, like a painting that stopped at the frame. Three. Check the scale of the detail. Real stones contain structure at every magnification. Cracks inside cracks. Inclusions. Tiny flaws no artist would think to add. Zoom in and the complexity keeps going. Generated imagery often runs out of detail. Past a certain zoom, it melts into smooth gradient soup. Four. Check the hands and the objects touching the stone. Contact is the hardest thing to fake. Weight, shadow, compression of skin, the way fingers curve around real mass. When a hand grips something heavy, the whole arm knows it. When a hand grips something that isn't there, there is always a tiny lie in the posture. Five. Check the audio. A real strike on real stone has a specific sound. Dense, uneven, with a tail that rings differently depending on what is inside. Foley artists can fake it, but they rarely fake all of it, all the time, perfectly synced to every impact. Six, and this is the one that matters most. Check the source. Who posted it. When. Where else does it appear. Is there a chain of custody, a mine, a seller, an auction record, a name you can look up. A miracle with no provenance is not a miracle. It is content. PART 8. WHY EVEN THE FAKE IS DOING SOMETHING REAL Here is where I'm going to disagree with the people who want to burn all of it down. Because a rendered gem is not worthless. It is doing real work on your nervous system, and that work deserves to be understood rather than sneered at. Humans have always built illusions to reach feelings that reality rations out too slowly. Theater. Cathedrals with stained glass. Fireworks. Cinema. Nobody calls a movie a lie because the spaceship wasn't real. The deal is clear. You know it is made. You consent to the feeling. The problem is never the illusion. The problem is the missing label. A simulation announced as a simulation is art. A simulation disguised as discovery is a weapon. Same pixels. Opposite ethics. The only thing that changes is whether you were told the truth before you felt the thing. So when a clip shows up carrying a title that admits what it is, that is not a trick. That is the rarest and most underrated move in modern media. Someone said the quiet part first. And the feeling you get afterward is cleaner for it, because you are not spending the next hour wondering if you were played. Remember that. In a feed full of costumes, the one wearing a name tag is the one worth trusting. PART 9. THE MONEY NOBODY MENTIONS Let's follow the cash, because it explains almost everything. A gemstone video earns attention. Attention earns reach. Reach earns three kinds of income. Direct. Ad revenue, subscriptions, sponsorships. The obvious pipe. Indirect. Traffic to a shop. A storefront selling polished stones, jewelry, kits, tools, mystery boxes. The video is a free sample. The checkout is the business. And third, the quiet one. Authority. Once you have a few million views, you are an expert, whether or not you ever held a real specimen. Authority converts into everything else. Now here is the part that should make your stomach tighten. Mystery boxes. Grab bags. Dig it yourself kits. Sealed stones sold with the promise that something spectacular might hide inside. Think about what that product is. It is the variable reward machine from earlier, sold back to you as a physical object, priced above its contents, with a promotional video showing someone else winning. The video shows the jackpot. The box contains the expected value. And the expected value of a cheap sealed rock is almost always a cheap rock. This is not an accusation against every seller. Plenty of honest dealers sell real stones with real descriptions at fair prices, and the best of them will happily tell you exactly what the odds are. But the structure rewards the opposite behavior, which is why the advice is simple. If the odds are hidden, assume they are bad. If the reveal is the product, assume the reveal is staged. If the price is a fraction of what the footage implies, assume the footage lies. PART 10. WHAT AN EXPERT ACTUALLY DOES Want to know how a real gemologist handles a mystery stone? Slowly. Boringly. With tools that make for terrible content. First, a loupe. Ten power magnification. They look for the structure of the surface, the pattern of the color, the telltale features that separate natural from lab. Second, a scale and a water test, to measure specific gravity. Real opal has a density in a known range. Plastic, glass, and some imitations land somewhere else entirely. Third, behavior under light. Ultraviolet. Polarized. Different materials answer these differently. A stone has a personality, and light is how you interview it. Fourth, the paper trail. Where was it found. Who sold it. What documents exist. In a serious market, a stone without a story is priced like a stone without a pedigree. Notice something. None of this is dramatic. None of it fits in forty seconds. None of it ends with a gasp. That is the entire problem and the entire lesson. Truth is slow. Attention is fast. The two are in a permanent tug of war, and whoever controls the clock controls the room. PART 11. THE CHEAPEST SUPERPOWER Let me give you the one habit that protects you from nearly everything in this post. When a clip makes you feel a surge, wait three seconds before you share it. That is it. Three seconds. Not to kill the feeling. Just to let the thinking part of your brain catch up with the feeling part. Ask two questions. What exactly am I being asked to believe? Who benefits if I believe it? If the answer to the first is something extraordinary, and the answer to the second is a person with a link in their bio, slow down. This is not paranoia. It is hygiene. The same way you wash your hands, not because every surface is deadly, but because the cost of doing it is nothing and the cost of not doing it is occasionally everything. You can still enjoy the show. You can still gasp. You just stop paying for it with your trust. PART 12. THE DEEPER THING Zoom out for a moment. We are living through the first era in history where seeing is no longer believing, and nobody has updated the instruction manual. For nearly all of human existence, an image was evidence. A photograph meant a thing happened. A video meant it happened in front of a lens. That rule held for roughly a hundred and fifty years, and entire legal systems, newsrooms, and personal relationships were built on top of it. That rule just expired. It did not expire with a bang. It expired the way a milk carton does. Quietly, in the dark, while everyone kept drinking. What replaces it? Not paranoia. A society that disbelieves everything is as broken as one that believes everything. Cynics are easy to fool too. They just get fooled in the other direction, dismissing real things because real things can now be dismissed. What replaces it is provenance. Receipts. Chains of custody. Systems that say who made this, with what, when, and who vouches for it. Cryptographic signatures at the moment of capture. Labels that survive a screenshot. Platforms that tell you plainly what is synthetic. We are building that. Slowly. Unevenly. Not fast enough. Until then, the responsibility falls where it always has, on the one person standing between a clip and a share button. You. PART 13. THE BEAUTIFUL PART I PROMISED Now the good news. I said this would turn beautiful, and I keep my promises. Strip away the money, the algorithm, the renders, the grifters. Underneath all of it sits a thing that is still true, still real, still waiting for you. The universe really does hide color inside ordinary things. It is not a trick. It is not marketing. It is the actual shape of reality. Plain gray granite carries minerals that glow under ultraviolet light like a hidden city. A dull pebble on a riverbank can be a fragment of a thing that took a hundred million years to form. A drop of water on a leaf bends light into a full spectrum, for free, every morning, to nobody in particular. The only reason a fake gemstone can move you is that a real one once moved somebody first. The counterfeit borrows its power from the original. Every render of a rainbow is a love letter to a rainbow that existed before the first camera. So here is the invitation. Go find the real thing. Not on a feed. In a museum with a velvet tray and a rock that is allowed to be dull. At a gem and mineral show where a grizzled dealer will hand you a lump and tell you what it is in four honest sentences. On a riverbank, with your own eyes, picking up stones and turning them in the sun. Hold a real opal under a lamp and rotate your wrist. Watch the color fall in and out of existence. Feel the small private shock of it. That shock cannot be rendered, because it is not in the pixels. It is in the fact that you are standing in the same universe as the thing, breathing the same air, aware that it took longer to make than your entire species has existed. That feeling is what the machine is trying to imitate. It is also what the machine can never deliver. PART 14. THE RULES, CONDENSED Screenshot this. Send it to the person in your family who forwards everything. One. Beauty is not evidence. The more perfect it looks, the more carefully you should look. Two. Color that doesn't move with the light is color that was never in the room. Three. A miracle without a source is just entertainment. Enjoy it as that. Four. If the odds are hidden, the odds are bad. Five. Labels are a form of respect. Trust the people who use them. Six. Wait three seconds. Ask who benefits. Seven. Support the slow experts. Miners, gemologists, dealers, museum curators, the people whose work doesn't fit in a clip. They are the reason any of this is real. Eight. Let yourself feel wonder anyway. Cynicism is not intelligence. Wonder with open eyes is the highest form of it. PART 15. WHAT HAPPENS NEXT Let me make a prediction, and you can come back and check me. Within a few years, the question will not be whether a clip like this is real. The question will be whether it is labeled. Real footage will carry cryptographic proof of origin, and its value will rise because of it. Synthetic footage will carry honest labels, and its value will hold because of that, as art, as entertainment, as a legitimate craft with real skill behind it. The dangerous category will be the middle. The unlabeled. The ambiguous. The thing that wants to be believed without earning it. Creators who label honestly will build the longest careers. Platforms that enforce labels will keep their users. Audiences who learn the tells will become impossible to exploit, and the exploiters will quietly move on to easier targets. The slow, truthful, boring work of verification will turn out to be the most valuable skill of the decade. You can start building it today, one three second pause at a time. PART 16. A SMALL EXPERIMENT FOR TONIGHT Try this before you sleep. Open the next clip that gives you that jolt. Before the reveal lands, say out loud what you expect to see. Then watch it. Then ask yourself what you actually felt, and where in your body. Most people discover the feeling arrives before the information does. The chest tightens a half second ahead of the facts. That gap is the whole game, and now that you know it exists, nobody can use it on you without your permission. Do it for a week. You will start noticing patterns everywhere, in headlines, in product launches, in the way a stranger tells a story. The trick is always the same. Build a gap. Fill it with feeling. Sell the feeling. You were never the audience. You were the product. Now you are the editor. THE LAST THING Let me bring this home. We began with a rock and a second you can't see inside it. Here is what that second really is. It is the gap between what you expect and what you get. It is the entire engine of curiosity. It is the reason anyone ever picked up a stone, cracked it, and looked. It is the oldest question our species knows how to ask: what is hiding in there? A civilization is the sum of everyone who refused to stop asking. The technology that fakes the answer has not killed the question. It has only made the question more valuable. Because now, when you finally do find something real, something unlabeled by any algorithm, unpolished by any editor, unrendered, unsponsored, and entirely indifferent to whether you are watching, it will mean more than it ever did. Real wonder is going to become scarce. And scarce things are the ones worth protecting. So protect it. Teach it. Share the real things generously and the fake things honestly. Keep your eyes open and your heart open and your thumb a little slower than the feed wants it to be. The stone is just the container. The color was always yours. The best feeds of the next decade will not be the ones that fool you best. They will be the ones that let you stay amazed without ever feeling cheated. Build for that. Follow that. Reward that. Everything else is just noise. If this changed how you'll watch the next one, repost the first line, and tell me one thing in the replies. What is the most beautiful real thing you have ever held in your hands? I'll read every answer.

Earth Unveiled

365,757 görüntüleme • 12 gün önce

Notch Clips. every other ad tool ignores the footage sitting in your dropbox and generates new clips from a prompt. notch opens your library instead. it watches every clip you have ever shot and understands what is inside each one. the mood. who is in frame. what they are doing. the quality. so when you ask it to build an ad, it does not start from scratch. it reaches into your past shoots, finds the clip that fits, and slots it into the exact moment that needs it. here is what that looks like. you had thirteen versions of one reaction shot. notch picked the one that worked and left the other twelve. not by filename. not by guessing from a reference image. it watched all thirteen and could tell the difference. that is the whole idea. the right clip was already in your library. every other tool was about to make a worse one. footage you already shot, turned into every ad you still need. drop your site and a few clips you have. notch builds the first ad from them.
0:40

Notch Clips. every other ad tool ignores the footage sitting in your dropbox and generates new clips from a prompt. notch opens your library instead. it watches every clip you have ever shot and understands what is inside each one. the mood. who is in frame. what they are doing. the quality. so when you ask it to build an ad, it does not start from scratch. it reaches into your past shoots, finds the clip that fits, and slots it into the exact moment that needs it. here is what that looks like. you had thirteen versions of one reaction shot. notch picked the one that worked and left the other twelve. not by filename. not by guessing from a reference image. it watched all thirteen and could tell the difference. that is the whole idea. the right clip was already in your library. every other tool was about to make a worse one. footage you already shot, turned into every ad you still need. drop your site and a few clips you have. notch builds the first ad from them.

Vinay Jain

407,810 görüntüleme • 5 ay önce

One of the largest trading firms in the world teaches new hires poker before it lets them near a book. A newspaper brought a camera to the table and sat down to play. The men across the felt are working Wall Street traders. The firm is Susquehanna, which built poker into its training programme decades ago and still runs it that way, on the argument that the card room teaches something no finance degree does. Gunjan Banerji from the Wall Street Journal plays the hands herself rather than interviewing them about it. A green table, a dealer, chips, 4 people who do this for a living. No lecture hall, no slides. The teaching happens between deals, while money is actually at stake. They break it into parts on camera. Risk management first. Then bet sizing. Then patience, which sounds like the soft one and is not. Then reading the person opposite when the only data available is how they behave with money on the table. The section on patience is the one most people skip. Folding is the correct action in the overwhelming majority of hands, and almost nobody can do it for hours without inventing a reason to play. The same failure shows up in a trading account as overtrading, and it kills more people than bad analysis. Then bet sizing, which is where the video earns the watch. Being right about the odds is the easy half. How much you put behind a correct read is the part that ends careers, and they work through it hand by hand instead of describing it in the abstract. The turn is what the traders admit about being wrong. A good decision loses regularly, a bad decision wins regularly, and the only way to last is to grade the process instead of the result. Everything else in the video sits downstream of that. It matters more now than when the firm started running these tables. Every model prices probability in a second and hands it to you for free. Nothing on the screen tells you how much of your account to put behind the number, or what to do after the number was right and you lost anyway. Free on YouTube, produced by a newspaper, filmed at a real table with real hands. The maths is public. The sizing is the job. 1 table. 4 traders. It is in the video.
16:30

One of the largest trading firms in the world teaches new hires poker before it lets them near a book. A newspaper brought a camera to the table and sat down to play. The men across the felt are working Wall Street traders. The firm is Susquehanna, which built poker into its training programme decades ago and still runs it that way, on the argument that the card room teaches something no finance degree does. Gunjan Banerji from the Wall Street Journal plays the hands herself rather than interviewing them about it. A green table, a dealer, chips, 4 people who do this for a living. No lecture hall, no slides. The teaching happens between deals, while money is actually at stake. They break it into parts on camera. Risk management first. Then bet sizing. Then patience, which sounds like the soft one and is not. Then reading the person opposite when the only data available is how they behave with money on the table. The section on patience is the one most people skip. Folding is the correct action in the overwhelming majority of hands, and almost nobody can do it for hours without inventing a reason to play. The same failure shows up in a trading account as overtrading, and it kills more people than bad analysis. Then bet sizing, which is where the video earns the watch. Being right about the odds is the easy half. How much you put behind a correct read is the part that ends careers, and they work through it hand by hand instead of describing it in the abstract. The turn is what the traders admit about being wrong. A good decision loses regularly, a bad decision wins regularly, and the only way to last is to grade the process instead of the result. Everything else in the video sits downstream of that. It matters more now than when the firm started running these tables. Every model prices probability in a second and hands it to you for free. Nothing on the screen tells you how much of your account to put behind the number, or what to do after the number was right and you lost anyway. Free on YouTube, produced by a newspaper, filmed at a real table with real hands. The maths is public. The sizing is the job. 1 table. 4 traders. It is in the video.

kompot

63,388 görüntüleme • 3 ay önce

The Measure of Governance: What a Number Can Say, and What We Must Do With It Every government says it is delivering. Between that claim and the citizen sits a number. In this clip I ask a plain question of the Ibrahim Index of African Governance: what is a number allowed to say? A number carries an authority its reader has done nothing to earn. That is its usefulness, and its danger. So before I repeat one, I put four questions to it. What was counted? What was left out? How were the parts combined? Which edition does it come from? Solomon Atah and I wrote “The Measure of Governance” together. It takes the Index apart, sets out what it can and cannot tell us, and asks what a country should do with its own number. No number will build a clinic or teach a child to read. People do that, in institutions that outlast them. Read the article: #IbrahimIndex #AfricanGovernance #MoIbrahimFoundation #AfricanLeadership #TheFutureIsNotAnAccident
3:08

The Measure of Governance: What a Number Can Say, and What We Must Do With It Every government says it is delivering. Between that claim and the citizen sits a number. In this clip I ask a plain question of the Ibrahim Index of African Governance: what is a number allowed to say? A number carries an authority its reader has done nothing to earn. That is its usefulness, and its danger. So before I repeat one, I put four questions to it. What was counted? What was left out? How were the parts combined? Which edition does it come from? Solomon Atah and I wrote “The Measure of Governance” together. It takes the Index apart, sets out what it can and cannot tell us, and asks what a country should do with its own number. No number will build a clinic or teach a child to read. People do that, in institutions that outlast them. Read the article: #IbrahimIndex #AfricanGovernance #MoIbrahimFoundation #AfricanLeadership #TheFutureIsNotAnAccident

Dr Malusi Gigaba

69,971 görüntüleme • 6 gün önce

context engineering vs graph engineering. every few months the list gets a new word and everyone treats it as a replacement for the last one. these two are not on the same list. one decides what the model sees this turn, the other decides what exists at all. the cleanest way to tell them apart is to ask what a single unit of work looks like. > context engineering is the window the window opens empty, every single time. you assemble what goes in it. the prompt, the docs, the history, the tool results. the assembling is the work. the window only grows. it never shrinks on its own, so eventually something gets dropped. usually from the middle. usually without telling you. then the turn ends and the window is thrown away. not archived, thrown away. the next turn opens empty again and you re-explain what you already explained. good context engineering is knowing what to leave out, not what to pack in. the unit of work is one window. > graph engineering is the structure the same material arrives from the same sources. instead of packing it into a window, you pull entities out of it, resolve the duplicates into one node, and write typed edges between them. nothing here is stored as text you hope to find again. it is stored as a thing with a name and its connections to other things. when the turn ends, the graph is still there. the next turn does not start from zero. it starts by querying what already exists, and the query walks edges instead of guessing at similarity. good graph engineering is deciding what counts as the same thing twice. the unit of work is one relationship. > they are not alternatives the graph is what refills the window. context engineering decides what fits. graph engineering decides what there is to choose from. remove the graph and every session starts blind. remove the context work and the best structure in the world arrives as an unreadable dump. that also tells you which one broke. the answer drifted from what you actually said, or forgot something from this same session. that is the window. the answer is coherent but invents a connection that does not exist, or cannot join two facts it has clearly seen. that is the structure. people debug the prompt because the prompt is the easiest thing to edit. it keeps taking the blame for failures that live a layer down. save this - then read the full breakdown below
0:40

context engineering vs graph engineering. every few months the list gets a new word and everyone treats it as a replacement for the last one. these two are not on the same list. one decides what the model sees this turn, the other decides what exists at all. the cleanest way to tell them apart is to ask what a single unit of work looks like. > context engineering is the window the window opens empty, every single time. you assemble what goes in it. the prompt, the docs, the history, the tool results. the assembling is the work. the window only grows. it never shrinks on its own, so eventually something gets dropped. usually from the middle. usually without telling you. then the turn ends and the window is thrown away. not archived, thrown away. the next turn opens empty again and you re-explain what you already explained. good context engineering is knowing what to leave out, not what to pack in. the unit of work is one window. > graph engineering is the structure the same material arrives from the same sources. instead of packing it into a window, you pull entities out of it, resolve the duplicates into one node, and write typed edges between them. nothing here is stored as text you hope to find again. it is stored as a thing with a name and its connections to other things. when the turn ends, the graph is still there. the next turn does not start from zero. it starts by querying what already exists, and the query walks edges instead of guessing at similarity. good graph engineering is deciding what counts as the same thing twice. the unit of work is one relationship. > they are not alternatives the graph is what refills the window. context engineering decides what fits. graph engineering decides what there is to choose from. remove the graph and every session starts blind. remove the context work and the best structure in the world arrives as an unreadable dump. that also tells you which one broke. the answer drifted from what you actually said, or forgot something from this same session. that is the window. the answer is coherent but invents a connection that does not exist, or cannot join two facts it has clearly seen. that is the structure. people debug the prompt because the prompt is the easiest thing to edit. it keeps taking the blame for failures that live a layer down. save this - then read the full breakdown below

Hanako

19,160 görüntüleme • 3 ay önce

One build now drives Blender from the inside, writes its own templates, and does in seconds what a modeller does in an afternoon It never exports a file and hands it over. It runs as an operator, calling the same functions a human clicks. > DRIVE - the build works through Blender’s own API, so every action is native and undoable > TEMPLATE - repeated structures turned into reusable node and modifier setups automatically > SCALE - one template applied across hundreds of variants in a single pass Generated meshes usually arrive as dead geometry. You get a shape and lose every parameter that produced it. Working inside Blender keeps the parameters alive. Change one value and the whole family of objects rebuilds itself. The templates are the real speedup. The build spots the pattern you keep repeating and turns it into a modifier stack. After that the hundredth variant costs the same as the second. The expensive part was always the setup, not the geometry. It also uses the features most people never touch, because geometry nodes punish anyone learning them under deadline. Nothing leaves the software. No round trips, no re-import, no fixing scale and orientation on the way back in. The operator, the template system and a full variant run are laid out below.
0:33

One build now drives Blender from the inside, writes its own templates, and does in seconds what a modeller does in an afternoon It never exports a file and hands it over. It runs as an operator, calling the same functions a human clicks. > DRIVE - the build works through Blender’s own API, so every action is native and undoable > TEMPLATE - repeated structures turned into reusable node and modifier setups automatically > SCALE - one template applied across hundreds of variants in a single pass Generated meshes usually arrive as dead geometry. You get a shape and lose every parameter that produced it. Working inside Blender keeps the parameters alive. Change one value and the whole family of objects rebuilds itself. The templates are the real speedup. The build spots the pattern you keep repeating and turns it into a modifier stack. After that the hundredth variant costs the same as the second. The expensive part was always the setup, not the geometry. It also uses the features most people never touch, because geometry nodes punish anyone learning them under deadline. Nothing leaves the software. No round trips, no re-import, no fixing scale and orientation on the way back in. The operator, the template system and a full variant run are laid out below.

wast3

40,481 görüntüleme • 26 gün önce

🚨 LIVE BASEBALL BROADCAST DERAILED AFTER “UNIDENTIFIED OBJECT” DROPS ONTO THE FIELD — “WE HAVE NO IDEA WHAT THIS IS.” Minor league game. Packed crowd. Cameras rolling at a Hub City Spartanburgers game. Something the length of a grown man’s leg touches down halfway between first and second base on the baseball field. It doesn’t bounce. It doesn’t roll. It floats. On air, the announcer calls it what everyone’s thinking: “An unidentified object.” “No one has any clue what it is.” They start guessing in real time: “Is that a foam finger that got caught in the rainstorm?” “Is it a balloon covered in dish soap?” “It looks like a giant ice cube you pulled out of your fridge…” For radio listeners, they estimate it’s about the length of one’s leg. It drifts behind the mound… slides toward third base foul territory… like it’s choosing its own path across the infield. Head groundskeeper runs out to scoop it up and even he struggles to contain it as it tries to lift back into the air. Thousands watching live in the stadium. Clips already spreading online. And the official description on a live baseball broadcast? “Unidentified object.” What do you think that was?
1:16

🚨 LIVE BASEBALL BROADCAST DERAILED AFTER “UNIDENTIFIED OBJECT” DROPS ONTO THE FIELD — “WE HAVE NO IDEA WHAT THIS IS.” Minor league game. Packed crowd. Cameras rolling at a Hub City Spartanburgers game. Something the length of a grown man’s leg touches down halfway between first and second base on the baseball field. It doesn’t bounce. It doesn’t roll. It floats. On air, the announcer calls it what everyone’s thinking: “An unidentified object.” “No one has any clue what it is.” They start guessing in real time: “Is that a foam finger that got caught in the rainstorm?” “Is it a balloon covered in dish soap?” “It looks like a giant ice cube you pulled out of your fridge…” For radio listeners, they estimate it’s about the length of one’s leg. It drifts behind the mound… slides toward third base foul territory… like it’s choosing its own path across the infield. Head groundskeeper runs out to scoop it up and even he struggles to contain it as it tries to lift back into the air. Thousands watching live in the stadium. Clips already spreading online. And the official description on a live baseball broadcast? “Unidentified object.” What do you think that was?

HustleBitch

163,451 görüntüleme • 8 ay önce