Загрузка видео...

Не удалось загрузить видео

На главную

Google Engineer: "I never think about a context window anymore" tasks run 36 hours. the window empties and refills the whole time, and he designed for that nothing lives in the prompt. the rules hit through tool outputs, exactly when needed "we want to make every mistake something that...

vartekxx's profile picture

vartekx

1,207 subscribers

83,504 просмотров • 1 месяц назад •via X (Twitter)

Наука и технологииОбразованиеНовости и политика
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Google engineer: "We don't prompt agents anymore. We build Loops and Graphs that prompt them for us." In 30 minutes he shows how they build agents at Google that do all the work, and why Loops are what make it possible. The missing piece most people skip: you're not supposed to talk to your agents. You're supposed to build the Loop that does. Watch it, then read the full guide on loops and graphs below
29:00

Google engineer: "We don't prompt agents anymore. We build Loops and Graphs that prompt them for us." In 30 minutes he shows how they build agents at Google that do all the work, and why Loops are what make it possible. The missing piece most people skip: you're not supposed to talk to your agents. You're supposed to build the Loop that does. Watch it, then read the full guide on loops and graphs below

Mahax

28,167 просмотров • 1 месяц назад

ANTHROPIC LEAKED 6 RULES THAT LET ONE FOLDER REPLACE A WHOLE AI ENGINEERING TEAM COSTING $3.2M A YEAR folder with 6 files replaced the AI engineers - and the team that spent 4 years writing them isn't needed anymore. file → commit → description matches → loads → runs → leaves a prompt has no history, a folder has a git log - one commit updates every session at once, and every edit shows up in the diff. two prompts fight over one window - two skills just both apply, and nobody merges anything by hand. 8 skills sit on disk, one enters the context - the other seven cost zero tokens until the description matches. a prompt can only ask - the folder carries build py and template.docx inside it, so the deterministic parts stay deterministic. the same skill works in Claude Code, in Cowork and through the API - a prompt lives only in the window you pasted it into. and the main one: a skill leaves the window when it's done - a prompt never leaves. save this - 6 rules, one folder, zero copy-paste ↓
0:59

ANTHROPIC LEAKED 6 RULES THAT LET ONE FOLDER REPLACE A WHOLE AI ENGINEERING TEAM COSTING $3.2M A YEAR folder with 6 files replaced the AI engineers - and the team that spent 4 years writing them isn't needed anymore. file → commit → description matches → loads → runs → leaves a prompt has no history, a folder has a git log - one commit updates every session at once, and every edit shows up in the diff. two prompts fight over one window - two skills just both apply, and nobody merges anything by hand. 8 skills sit on disk, one enters the context - the other seven cost zero tokens until the description matches. a prompt can only ask - the folder carries build py and template.docx inside it, so the deterministic parts stay deterministic. the same skill works in Claude Code, in Cowork and through the API - a prompt lives only in the window you pasted it into. and the main one: a skill leaves the window when it's done - a prompt never leaves. save this - 6 rules, one folder, zero copy-paste ↓

Sprytix

11,389 просмотров • 2 месяцев назад

Google Brain founder, Andrew Ng: "One of the most important skills of the future is the ability to tell a computer exactly what you want so it can do it for you." In this talk, he explains why the biggest opportunity isn't just using AI. It's learning how to turn work into agentic workflows, break tasks into steps, give agents the right tools and context, add evals, and make the whole system reliable. That's the difference between prompting an AI and engineering an agent that actually works. Watch it today, then read the full guide in the article below.
26:58

Google Brain founder, Andrew Ng: "One of the most important skills of the future is the ability to tell a computer exactly what you want so it can do it for you." In this talk, he explains why the biggest opportunity isn't just using AI. It's learning how to turn work into agentic workflows, break tasks into steps, give agents the right tools and context, add evals, and make the whole system reliable. That's the difference between prompting an AI and engineering an agent that actually works. Watch it today, then read the full guide in the article below.

Rahul

14,431 просмотров • 2 месяцев назад

you're running /compact every 10 minutes to fix a problem that disappears the second you give Claude one job instead of five. Anthropic shipped a clean 4-minute video on context management. /compact when the window fills up. /clear to reset. small files over big files. the practical tips work. the video covers the cleanup. it doesn't cover why your window keeps filling up in the first place, and the answer is usually that you're holding too many jobs in one prompt. look at what most people put in a single Claude call: > guess the right answer > choose between options > set the position size > watch the exit window > read the news for sentiment five tools' worth of state in one conversation. of course the window fills. i ran into this on a Polymarket bot: the prediction-layer version ate context constantly. /compact every 8-10 turns. i thought i needed better memory hygiene. i needed fewer jobs per call. i stripped the prompt down to one job: veto an entry i had already decided to take. the context window stopped filling. /compact went from every 8 turns to once a session, if ever. over 90 days that prompt blocked 128 trades. watch the 4 min, then ask why your window keeps needing it. clean context. lean prompt. nothing to compact.
3:26

you're running /compact every 10 minutes to fix a problem that disappears the second you give Claude one job instead of five. Anthropic shipped a clean 4-minute video on context management. /compact when the window fills up. /clear to reset. small files over big files. the practical tips work. the video covers the cleanup. it doesn't cover why your window keeps filling up in the first place, and the answer is usually that you're holding too many jobs in one prompt. look at what most people put in a single Claude call: > guess the right answer > choose between options > set the position size > watch the exit window > read the news for sentiment five tools' worth of state in one conversation. of course the window fills. i ran into this on a Polymarket bot: the prediction-layer version ate context constantly. /compact every 8-10 turns. i thought i needed better memory hygiene. i needed fewer jobs per call. i stripped the prompt down to one job: veto an entry i had already decided to take. the context window stopped filling. /compact went from every 8 turns to once a session, if ever. over 90 days that prompt blocked 128 trades. watch the 4 min, then ask why your window keeps needing it. clean context. lean prompt. nothing to compact.

Mnimiy

20,635 просмотров • 4 месяцев назад

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 просмотров • 2 месяцев назад

Ox Alpha is wired into codex now, and it's free for the whole week!! no api key, no billing. It sits in the picker next to the models we already pay for, and our native models never moved. Vision and a million tokens of context, costing nothing until the window shuts. so we stopped picking the files we think matter and started handing it the whole repo. Seven days is the whole deal. after that it goes back to being something we'd have to pay for, so this week is for the jobs that are usually too big. Here's the whole setup👇
0:24

Ox Alpha is wired into codex now, and it's free for the whole week!! no api key, no billing. It sits in the picker next to the models we already pay for, and our native models never moved. Vision and a million tokens of context, costing nothing until the window shuts. so we stopped picking the files we think matter and started handing it the whole repo. Seven days is the whole deal. after that it goes back to being something we'd have to pay for, so this week is for the jobs that are usually too big. Here's the whole setup👇

Ziwen

271,047 просмотров • 1 месяц назад

SpaceXAI engineer (ex-Cursor): "I run 20+ GrokBot agents and I haven't written a prompt in months. They run on loops now I delegate the delegation - one Chief of Staff drives the other 20 and I only see what needs a decision" In a 29-min session she walks the whole shift: the loop, the roster, and where the human still sits this is worth more than 10 hours of paid courses watch it today, then read the full ruleset in the article below ↓
29:15

SpaceXAI engineer (ex-Cursor): "I run 20+ GrokBot agents and I haven't written a prompt in months. They run on loops now I delegate the delegation - one Chief of Staff drives the other 20 and I only see what needs a decision" In a 29-min session she walks the whole shift: the loop, the roster, and where the human still sits this is worth more than 10 hours of paid courses watch it today, then read the full ruleset in the article below ↓

Carnage

34,524 просмотров • 1 месяц назад

Anthropic engineer: "You're not supposed to prompt Claude. You're supposed to build a system that prompts itself." this is one of the best workflows I've seen in a long time in this video she breaks down exactly how most people are using Claude: - the 14% you lose to CLAUDE.md before typing a word - the automation workflows most users don't know exist - the daily task pipelines that run without touching the keyboard - the daily workflows Anthropic's own engineers automated first if you've been using Claude for more than a month and never left the chat window, you've been using one agent when you could be running a team of them instead of another show tonight, watch this make sure to bookmark it before it gets lost in your feed the guide is in the article below
29:47

Anthropic engineer: "You're not supposed to prompt Claude. You're supposed to build a system that prompts itself." this is one of the best workflows I've seen in a long time in this video she breaks down exactly how most people are using Claude: - the 14% you lose to CLAUDE.md before typing a word - the automation workflows most users don't know exist - the daily task pipelines that run without touching the keyboard - the daily workflows Anthropic's own engineers automated first if you've been using Claude for more than a month and never left the chat window, you've been using one agent when you could be running a team of them instead of another show tonight, watch this make sure to bookmark it before it gets lost in your feed the guide is in the article below

Khairallah AL-Awady

923,831 просмотров • 4 месяцев назад

I tried to reverse engineer a prompt for this game. Used that in Cursor, with NO Ultracode/subagents only Opus 5 high and this is the result! The repo and the prompt are below
1:20

I tried to reverse engineer a prompt for this game. Used that in Cursor, with NO Ultracode/subagents only Opus 5 high and this is the result! The repo and the prompt are below

Leon Lin

43,161 просмотров • 2 месяцев назад

Introducing PAI 1.0: The First Version of Personal Super Intelligence PAI is the world's first context-aware, widget based AI agent that floats over your screen to read your context in real time and proactively suggest next steps, ensuring safety by executing actions only with your explicit approval. The floating AI interface designed to eliminate the prompt inequality and barrier. Structurally eliminating the "navigate to a chat window, write a prompt, explain your context" workflow that existing AI services have treated as a given for years.
8:24

Introducing PAI 1.0: The First Version of Personal Super Intelligence PAI is the world's first context-aware, widget based AI agent that floats over your screen to read your context in real time and proactively suggest next steps, ensuring safety by executing actions only with your explicit approval. The floating AI interface designed to eliminate the prompt inequality and barrier. Structurally eliminating the "navigate to a chat window, write a prompt, explain your context" workflow that existing AI services have treated as a given for years.

Eon (Yulheon) Seung

131,471 просмотров • 4 месяцев назад

NVIDIA CEO Jensen Huang: "Nobody is just writing prompts anymore, the new job is building and running loops" He says this is the shift that defines the rest of 2026 In 23 minutes, he explains why the best engineers are moving from prompting to looping Prompt Output Check Fix Run again That is the new workflow Bookmark and watch the interview Then save the full framework below
23:23

NVIDIA CEO Jensen Huang: "Nobody is just writing prompts anymore, the new job is building and running loops" He says this is the shift that defines the rest of 2026 In 23 minutes, he explains why the best engineers are moving from prompting to looping Prompt Output Check Fix Run again That is the new workflow Bookmark and watch the interview Then save the full framework below

rari

169,874 просмотров • 3 месяцев назад

Сreator of Claude Code: BORIS CHERNY, WHO BUILT CLAUDE CODE, SAID IT PLAINLY." I don't prompt Claude anymore. I have loops running that prompt Claude and figure out what to do. My job is to write loops. "He is not typing anymore. He runs loops that write the code for him. That is the shift most people miss. Not a better prompt. A loop that keeps running while you are gone.People sell $500 courses on this. The person who built the tool said it for free. The exact setup is in the article below.Bookmark this. Follow CyrilXBT
18:08

Сreator of Claude Code: BORIS CHERNY, WHO BUILT CLAUDE CODE, SAID IT PLAINLY." I don't prompt Claude anymore. I have loops running that prompt Claude and figure out what to do. My job is to write loops. "He is not typing anymore. He runs loops that write the code for him. That is the shift most people miss. Not a better prompt. A loop that keeps running while you are gone.People sell $500 courses on this. The person who built the tool said it for free. The exact setup is in the article below.Bookmark this. Follow CyrilXBT

CyrilXBT

38,827 просмотров • 3 месяцев назад

"100 million words context window is already possible, which is roughly what a human hears in a lifetime. Inference support is the only bottleneck to achieve it. And AI Models actually do learn during the context window, without changing the weights." ~ Anthropic CEO Dario Amodei (On the 2nd point, there was this brilliant Google Paper published last week that says, LLMs can learn in context from examples in the prompt, can pick up new patterns while answering, yet their stored weights never change.) --- From 'Alex Kantrowitz' YT Channel (Full Video link in comment)
0:59

"100 million words context window is already possible, which is roughly what a human hears in a lifetime. Inference support is the only bottleneck to achieve it. And AI Models actually do learn during the context window, without changing the weights." ~ Anthropic CEO Dario Amodei (On the 2nd point, there was this brilliant Google Paper published last week that says, LLMs can learn in context from examples in the prompt, can pick up new patterns while answering, yet their stored weights never change.) --- From 'Alex Kantrowitz' YT Channel (Full Video link in comment)

Rohan Paul

631,729 просмотров • 1 год назад

Joe Rogan “This is my take on this whole Make America Great Again thing” “Where the f*ck did we come up with a hundred and whatever billion dollars to send to Ukraine and we don't have any money to try to do something about these insanely impoverished, crime-ridden, gang-ridden, drug-ridden communities? We don't do anything? We have nothing? I mean, this is my take on this whole Make America Great Again thing: You want to Make America Great Again? Make it so there's less losers. Make it so that more people have a f*cking chance. The idea that everyone starts on the same line, I mean, I'm not talking about equality of outcome. That's not possible. But equality of opportunity is possible. That's a possible goal. And at least we could advance that. At least we could do something to just change the course of who knows how many people's lives. And we don't do a f*cking thing about it.“
1:07

Joe Rogan “This is my take on this whole Make America Great Again thing” “Where the f*ck did we come up with a hundred and whatever billion dollars to send to Ukraine and we don't have any money to try to do something about these insanely impoverished, crime-ridden, gang-ridden, drug-ridden communities? We don't do anything? We have nothing? I mean, this is my take on this whole Make America Great Again thing: You want to Make America Great Again? Make it so there's less losers. Make it so that more people have a f*cking chance. The idea that everyone starts on the same line, I mean, I'm not talking about equality of outcome. That's not possible. But equality of opportunity is possible. That's a possible goal. And at least we could advance that. At least we could do something to just change the course of who knows how many people's lives. And we don't do a f*cking thing about it.“

Wall Street Apes

2,049,676 просмотров • 2 лет назад

PRESIDENT TRUMP IS READY TO MAKE THE FINAL MOVE 👀 THE PIECES ARE ALLINGED FOR THE FIRST TIME SINCE 2017 HE KNOWS THE WINDOW IS CLOSING FAST AND WILL NEVER OPEN AGAIN PRESIDENT TRUMP IS ABOUT TO… (Read more in the first comment)
0:22

PRESIDENT TRUMP IS READY TO MAKE THE FINAL MOVE 👀 THE PIECES ARE ALLINGED FOR THE FIRST TIME SINCE 2017 HE KNOWS THE WINDOW IS CLOSING FAST AND WILL NEVER OPEN AGAIN PRESIDENT TRUMP IS ABOUT TO… (Read more in the first comment)

Matt Wallace

582,184 просмотров • 7 месяцев назад

Zohran Mamdani on how this NYC election impacts the arts and music scene: “If we want to make the arts something that every person can pursue — not just a luxury for the rich — then we have to ensure that the artists can afford to live in that same city.”
0:32

Zohran Mamdani on how this NYC election impacts the arts and music scene: “If we want to make the arts something that every person can pursue — not just a luxury for the rich — then we have to ensure that the artists can afford to live in that same city.”

Pop Base

167,852 просмотров • 11 месяцев назад

WHAT THE F*CK. SpaceXAI lead engineer (ex-Cursor): "99% of people are still typing into a chat window. 1% have already hired a team I run 6 bots. My Chief of Staff wakes me at 5am with my whole day planned, and my travel bot hands work to my flight bot and my hotel bot while I sleep." In 27 minutes, he walks through how he built that team from zero. Chat → Copilot → Grok Bot → Bot Teams → Self-Improving Systems Prompting was the old way to work. Hiring is the new one. This talk beats most $1,000 agent courses. Bookmark it and watch today. Then read how to hire your first Grok Bot in the article below ↓
26:38

WHAT THE F*CK. SpaceXAI lead engineer (ex-Cursor): "99% of people are still typing into a chat window. 1% have already hired a team I run 6 bots. My Chief of Staff wakes me at 5am with my whole day planned, and my travel bot hands work to my flight bot and my hotel bot while I sleep." In 27 minutes, he walks through how he built that team from zero. Chat → Copilot → Grok Bot → Bot Teams → Self-Improving Systems Prompting was the old way to work. Hiring is the new one. This talk beats most $1,000 agent courses. Bookmark it and watch today. Then read how to hire your first Grok Bot in the article below ↓

GenAI

53,217 просмотров • 8 дней назад

Elon Musk sat in the Oval Office and told the President of the United States exactly what was coming. Nothing happened. Musk: “The one thing we really need to do is set up an AI regulatory agency. You don’t just come shooting from the hip throwing out regulations. You just start with insight.” Not control. Not panic. Insight. Walk in. Learn what the companies are building. Then write rules that everyone agrees to follow. The same way the NFL sets the rules before anyone steps on the field. Obama nodded. Then did absolutely nothing. Musk: “People didn’t realize where AI was headed at that time. AI seemed like some super futuristic sci-fi.” The most powerful government on earth was handed the blueprint. They read it like a movie plot. That misread has a price. We are nowhere close to finishing paying it. While Washington waited for AI to look like something they recognized, the builders never stopped building. Now Musk: “This is gonna be smarter than the smartest human. And my predictions are coming absolutely true.” Sit with that. For the entire stretch of recorded history, the smartest person in the room set the ceiling. That ceiling is about to become a floor. We are building something that will look back at the most brilliant human mind the way we look at a calculator. The window was open. It needed one conversation. One decision. One agency that started with the word insight and nothing else. That window closed. The game is already being played. Nobody agreed to the rules. The referee never showed up.
1:00

Elon Musk sat in the Oval Office and told the President of the United States exactly what was coming. Nothing happened. Musk: “The one thing we really need to do is set up an AI regulatory agency. You don’t just come shooting from the hip throwing out regulations. You just start with insight.” Not control. Not panic. Insight. Walk in. Learn what the companies are building. Then write rules that everyone agrees to follow. The same way the NFL sets the rules before anyone steps on the field. Obama nodded. Then did absolutely nothing. Musk: “People didn’t realize where AI was headed at that time. AI seemed like some super futuristic sci-fi.” The most powerful government on earth was handed the blueprint. They read it like a movie plot. That misread has a price. We are nowhere close to finishing paying it. While Washington waited for AI to look like something they recognized, the builders never stopped building. Now Musk: “This is gonna be smarter than the smartest human. And my predictions are coming absolutely true.” Sit with that. For the entire stretch of recorded history, the smartest person in the room set the ceiling. That ceiling is about to become a floor. We are building something that will look back at the most brilliant human mind the way we look at a calculator. The window was open. It needed one conversation. One decision. One agency that started with the word insight and nothing else. That window closed. The game is already being played. Nobody agreed to the rules. The referee never showed up.

Dustin

31,091 просмотров • 6 месяцев назад

Impeccable creator Paul Bakaus says AI should be cognitive delegation, not cognitive surrender: "Cognitive delegation is, I'm using Google Maps, and Google Maps tells me where I can go as quickly as possible." "But what if I let Google Maps decide where I want to go to? Now I kind of cognitive surrender to the application, and I think that's very true for LLMs if you're not careful." "You prompt something, and then it creates a beautiful plan, and then it's eight pages long. You're not going to read through that plan. You're going to scroll through that plan and skim it, and it's like, 'I guess the model knows what it's doing.'" "I think delegating to the model is great, but preserving that point of view, making sure that you're still the one driving is also really important." Paul Bakaus
1:49

Impeccable creator Paul Bakaus says AI should be cognitive delegation, not cognitive surrender: "Cognitive delegation is, I'm using Google Maps, and Google Maps tells me where I can go as quickly as possible." "But what if I let Google Maps decide where I want to go to? Now I kind of cognitive surrender to the application, and I think that's very true for LLMs if you're not careful." "You prompt something, and then it creates a beautiful plan, and then it's eight pages long. You're not going to read through that plan. You're going to scroll through that plan and skim it, and it's like, 'I guess the model knows what it's doing.'" "I think delegating to the model is great, but preserving that point of view, making sure that you're still the one driving is also really important." Paul Bakaus

a16z

80,185 просмотров • 3 месяцев назад