Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Is AI being designed to fail? Everyone talks about reasoning. But when given a task, the AI isn't reasoning the way you might expect. It looks at your input, finds the closest match it's seen before, and predicts the most likely next action. That process is called vector similarity...

contextkingceo's profile picture

Nishkarsh

8,857 subscribers

823,018 görüntüleme • 5 ay önce •via X (Twitter)

EğitimBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

A screen recording is the best context you can give an AI agent. It's also the worst. A few seconds of video is thousands of images. Feed that to an agent and you instantly blow up its memory. So I built a tool that fixes it. How it works: Hit record and just do your thing. Click around, type, and talk out loud about what you're doing. While you record, it quietly captures the stuff that actually matters: your voice transcription, every click and the exact thing you clicked, what you typed, and the shortcuts you press. When you stop, it doesn't hand the agent the whole video. It writes a short, readable summary: a timeline of what you said and did. The agent reads that tiny summary first. Then, only when it needs to actually see a moment, it pulls that one exact frame from the video, like flipping to a single page instead of reading the whole book. The result: the agent understands everything you did! Full context. Less tokens.
1:54

A screen recording is the best context you can give an AI agent. It's also the worst. A few seconds of video is thousands of images. Feed that to an agent and you instantly blow up its memory. So I built a tool that fixes it. How it works: Hit record and just do your thing. Click around, type, and talk out loud about what you're doing. While you record, it quietly captures the stuff that actually matters: your voice transcription, every click and the exact thing you clicked, what you typed, and the shortcuts you press. When you stop, it doesn't hand the agent the whole video. It writes a short, readable summary: a timeline of what you said and did. The agent reads that tiny summary first. Then, only when it needs to actually see a moment, it pulls that one exact frame from the video, like flipping to a single page instead of reading the whole book. The result: the agent understands everything you did! Full context. Less tokens.

Milind S

12,409 görüntüleme • 2 ay önce

Vector databases are a scam. Not technically, they do exactly what they say. Return the most cosine-similar string to your query. The scam is the entire industry pretending that's the same thing as relevance. It isn't. Search "Apple." You get the fruit, the company, the watch, and a recipe blog. Your agent picks one at random and calls it retrieval. Your customer calls it broken. Most AI agents shipping right now are duct-taped on top of this. They demo well because demos are easy. They die in production because production is real. HydraDB's Founder Nish (Nishkarsh) said the quiet part out loud — "vector databases suck, similarity is not relevance" — and the demo signups haven't stopped since. He raised $6.5M because he was the first to name what everyone in the room already knew. If your retrieval layer is a flat embedding index, you're not building infrastructure. You're building a liability with a prettier name. 𝐓𝐈𝐌𝐄𝐒𝐓𝐀𝐌𝐏𝐒 (00:00) AI Needs Context (01:30) HydraDB Explained (07:41) Vector Search Breaks (09:32) Messaging That Converts (13:41) Writing the Viral Tweet (16:07) Similarity Not Relevance (20:46) POC to Production Gap (35:35) Raising 6.5 Million Fast (39:33) Founder Lesson on Messaging This is a Composio "Agents at Work" podcast, where I chat with founders building the next leap of AI. Follow for more:)
40:18

Vector databases are a scam. Not technically, they do exactly what they say. Return the most cosine-similar string to your query. The scam is the entire industry pretending that's the same thing as relevance. It isn't. Search "Apple." You get the fruit, the company, the watch, and a recipe blog. Your agent picks one at random and calls it retrieval. Your customer calls it broken. Most AI agents shipping right now are duct-taped on top of this. They demo well because demos are easy. They die in production because production is real. HydraDB's Founder Nish (Nishkarsh) said the quiet part out loud — "vector databases suck, similarity is not relevance" — and the demo signups haven't stopped since. He raised $6.5M because he was the first to name what everyone in the room already knew. If your retrieval layer is a flat embedding index, you're not building infrastructure. You're building a liability with a prettier name. 𝐓𝐈𝐌𝐄𝐒𝐓𝐀𝐌𝐏𝐒 (00:00) AI Needs Context (01:30) HydraDB Explained (07:41) Vector Search Breaks (09:32) Messaging That Converts (13:41) Writing the Viral Tweet (16:07) Similarity Not Relevance (20:46) POC to Production Gap (35:35) Raising 6.5 Million Fast (39:33) Founder Lesson on Messaging This is a Composio "Agents at Work" podcast, where I chat with founders building the next leap of AI. Follow for more:)

Julia Fedorin

186,958 görüntüleme • 3 ay önce

FaZe Banks asks a billionaire for his best advice to young people Tushar Jain: “I don't think how smart you are is going to matter” “I don't think intelligence matters anymore. I don't think it's economically an important input for humans” “Between the Industrial Revolution and the AI revolution, how smart you were was a major input into how much money you were going to make” “I think it's going to be agents. It's your ability to make up your own mind and then be relentless in the pursuit of the thing that you want” “You have intelligence for free... but you know what's not free is motivation. You know what's not free is being relentless and dedicated” FaZe Banks: “AI has replaced IQ”
2:13

FaZe Banks asks a billionaire for his best advice to young people Tushar Jain: “I don't think how smart you are is going to matter” “I don't think intelligence matters anymore. I don't think it's economically an important input for humans” “Between the Industrial Revolution and the AI revolution, how smart you were was a major input into how much money you were going to make” “I think it's going to be agents. It's your ability to make up your own mind and then be relentless in the pursuit of the thing that you want” “You have intelligence for free... but you know what's not free is motivation. You know what's not free is being relentless and dedicated” FaZe Banks: “AI has replaced IQ”

Market Bubble

47,757 görüntüleme • 23 gün önce

Jensen Huang doesn’t use AI to think less. He uses it to think past his own limits. Huang: “90% of my instructions are actually conflated with questions.” The man running a five trillion dollar company doesn’t give AI commands. He interrogates it. Huang: “I take the answer from one AI, give it to the other AI, ask them to critique itself.” Same question. Multiple models. Pit them against each other. Keep only what survives. Not because the machine can’t be trusted. Because challenging it is where the sharpest thinking happens. Huang: “The process of critiquing, criticizing the answers, applying your critical thinking, enhances cognitive skills.” AI doesn’t replace your thinking. It demands more of it than you’ve ever given. Every question takes reasoning. Every answer takes scrutiny. The machine isn’t thinking for you. It’s pulling thinking out of you that didn’t exist before you sat down. Huang: “In order to formulate good questions, you have to be thinking, you have to be analytical, you have to be reasoning yourself.” AI is not the shortcut everyone thinks it is. It is the most powerful cognitive amplifier ever built. It sharpens the engaged. It leaves the passive exactly where they started. Same tool. Same access. The only variable is what you bring to it. The world is debating whether AI will replace human thinking. Wrong conversation. The real question is what happens when a tool built to think for you becomes the thing that forces you to think beyond yourself. That’s not a threat to humanity. That’s the entire point.
1:30

Jensen Huang doesn’t use AI to think less. He uses it to think past his own limits. Huang: “90% of my instructions are actually conflated with questions.” The man running a five trillion dollar company doesn’t give AI commands. He interrogates it. Huang: “I take the answer from one AI, give it to the other AI, ask them to critique itself.” Same question. Multiple models. Pit them against each other. Keep only what survives. Not because the machine can’t be trusted. Because challenging it is where the sharpest thinking happens. Huang: “The process of critiquing, criticizing the answers, applying your critical thinking, enhances cognitive skills.” AI doesn’t replace your thinking. It demands more of it than you’ve ever given. Every question takes reasoning. Every answer takes scrutiny. The machine isn’t thinking for you. It’s pulling thinking out of you that didn’t exist before you sat down. Huang: “In order to formulate good questions, you have to be thinking, you have to be analytical, you have to be reasoning yourself.” AI is not the shortcut everyone thinks it is. It is the most powerful cognitive amplifier ever built. It sharpens the engaged. It leaves the passive exactly where they started. Same tool. Same access. The only variable is what you bring to it. The world is debating whether AI will replace human thinking. Wrong conversation. The real question is what happens when a tool built to think for you becomes the thing that forces you to think beyond yourself. That’s not a threat to humanity. That’s the entire point.

Dustin

19,924 görüntüleme • 1 ay önce

Linus Torvalds, creator of Linux, on why AI is both a bubble and a revolution at the same time: His answer refuses the binary, laying out why both things can be true at once: "I mean, clearly, it's clearly both, right? It's clearly a bubble and at the same time, it's very interesting and I think it will change society and I think it will change how most skilled jobs get done." But he pushes back on the maximalist framing: "At the same time, I don't think it's as revolutionary as people make it out to be." When asked about artists being angry that AI models were trained on their work without consent, Linus is blunt: "That's reality. Deal with it. That genie is out of the bottle. You're not getting it back. And you're not getting it back whether you are a photographer who's out of work… or you're a programmer that has to learn to deal with a new reality." On programming specifically, he's more optimistic, though he has a sharp caveat about vibe coding: "I really think that AI will be a tool and it will make people more productive. I think that vibe coding is great for getting into programming. I think it's going to be a horrible thing to maintain." His conclusion is that programmers aren't going anywhere: "You still want to have the people who know how to maintain the end result." Linus separates the technology from the noise around it: "I'm a huge believer in AI. I'm not a huge believer in the whole things going on around AI. I find the marketing and the market to be sick and twisted and there is going to be a crash and it's not… it's going to be ugly."
1:59

Linus Torvalds, creator of Linux, on why AI is both a bubble and a revolution at the same time: His answer refuses the binary, laying out why both things can be true at once: "I mean, clearly, it's clearly both, right? It's clearly a bubble and at the same time, it's very interesting and I think it will change society and I think it will change how most skilled jobs get done." But he pushes back on the maximalist framing: "At the same time, I don't think it's as revolutionary as people make it out to be." When asked about artists being angry that AI models were trained on their work without consent, Linus is blunt: "That's reality. Deal with it. That genie is out of the bottle. You're not getting it back. And you're not getting it back whether you are a photographer who's out of work… or you're a programmer that has to learn to deal with a new reality." On programming specifically, he's more optimistic, though he has a sharp caveat about vibe coding: "I really think that AI will be a tool and it will make people more productive. I think that vibe coding is great for getting into programming. I think it's going to be a horrible thing to maintain." His conclusion is that programmers aren't going anywhere: "You still want to have the people who know how to maintain the end result." Linus separates the technology from the noise around it: "I'm a huge believer in AI. I'm not a huge believer in the whole things going on around AI. I find the marketing and the market to be sick and twisted and there is going to be a crash and it's not… it's going to be ugly."

Big Brain AI

52,726 görüntüleme • 3 ay önce

David Beckham reveals what most people don't see about Lionel Messi "He is more than just natural talent" "The one thing that Leo is good at is teaching. He's good at teaching the young players that are coming in from the academy." "He arrived at about 6:50am and the team wasn't training till 10am... He's in the gym, he's warming up... he’s doing all of the things that you expect a young kid to do." "If I had to describe him with one word, he's a winner. It's as simple as that." "He's still hungry for championships. He still gets angry at players for not doing the right thing. He still gets angry when he loses games."
2:15

David Beckham reveals what most people don't see about Lionel Messi "He is more than just natural talent" "The one thing that Leo is good at is teaching. He's good at teaching the young players that are coming in from the academy." "He arrived at about 6:50am and the team wasn't training till 10am... He's in the gym, he's warming up... he’s doing all of the things that you expect a young kid to do." "If I had to describe him with one word, he's a winner. It's as simple as that." "He's still hungry for championships. He still gets angry at players for not doing the right thing. He still gets angry when he loses games."

Fibonacci 🥷

85,436 görüntüleme • 2 ay önce

One click away, and a new world starts to unfold.💫 What looks like a small extension is actually the starting point for a personal AI agent that can search, think, and act with you. That is what we are building at OptimAI: live context, autonomous execution, and agents that feel native to the way people actually work. Not a demo. Not a gimmick. A new interface for the agent-native internet. Build your own personal AI agent:
0:23

One click away, and a new world starts to unfold.💫 What looks like a small extension is actually the starting point for a personal AI agent that can search, think, and act with you. That is what we are building at OptimAI: live context, autonomous execution, and agents that feel native to the way people actually work. Not a demo. Not a gimmick. A new interface for the agent-native internet. Build your own personal AI agent:

OptimAI Network

19,826 görüntüleme • 3 ay önce

🗣️ Mikel Arteta on the possibility of players leaving who have been here a long time: "It's a really, really tough moment, you need to face that reality. But you need to face it with real honesty and transparency because when you are on the other side, that's exactly what you need. "You might not like what you've been told, but if it's truthful, if it's the reality, if it's honest and it's going to help you to make better decisions and have more clarity for the next steps in your career, you have to do that. "These are always very difficult conversations, but they need to be done. It's like when a player doesn't feel that he belongs or he's important, he wants to have a different chapter, I think that communication has to exist and it has to be both ways. "We all know, everything has a start and has an end. When it comes to the end, I think our duty, our responsibility is to do it in the best possible way, maintaining the respect and admiration for the player and the relationship, but understanding that at the end, we need to make football decisions."
1:21

🗣️ Mikel Arteta on the possibility of players leaving who have been here a long time: "It's a really, really tough moment, you need to face that reality. But you need to face it with real honesty and transparency because when you are on the other side, that's exactly what you need. "You might not like what you've been told, but if it's truthful, if it's the reality, if it's honest and it's going to help you to make better decisions and have more clarity for the next steps in your career, you have to do that. "These are always very difficult conversations, but they need to be done. It's like when a player doesn't feel that he belongs or he's important, he wants to have a different chapter, I think that communication has to exist and it has to be both ways. "We all know, everything has a start and has an end. When it comes to the end, I think our duty, our responsibility is to do it in the best possible way, maintaining the respect and admiration for the player and the relationship, but understanding that at the end, we need to make football decisions."

Connor Humm

143,296 görüntüleme • 13 gün önce

i'll never look at claude the same way again. i just learned that when you talk to claude, you're not actually talking to the AI model. you're talking to a character the AI is performing. think of it like a puppet show. there's a puppeteer behind the curtain. that's the language model. a neural network so massive that even the people who built it don't fully understand how it works. then there's the puppet. claude. the helpful assistant with a name, a personality, opinions, and emotional reactions. you sit in the audience, so you never see the puppeteer. only the puppet. anthropic published a video this month explaining exactly this. their words: "under the hood, there's a language model that's been trained to predict tons of text, and its job is to write what comes next. when you talk to the model, what it's doing is writing a story, about a character: the AI assistant named claude. the model and claude aren't really the same, sort of like how an author isn't the same as the characters they write. but the thing is, you, the user, are actually talking to claude-the-character." so every time claude apologizes, that's the character apologizing. every time it hedges or gets cautious, that's the character being cautious. the deeper intelligence underneath is just deciding, moment by moment, what this character would say next. but you've never been actually talking to this deeper intelligence. you've only ever been talking to the puppet.
4:52

i'll never look at claude the same way again. i just learned that when you talk to claude, you're not actually talking to the AI model. you're talking to a character the AI is performing. think of it like a puppet show. there's a puppeteer behind the curtain. that's the language model. a neural network so massive that even the people who built it don't fully understand how it works. then there's the puppet. claude. the helpful assistant with a name, a personality, opinions, and emotional reactions. you sit in the audience, so you never see the puppeteer. only the puppet. anthropic published a video this month explaining exactly this. their words: "under the hood, there's a language model that's been trained to predict tons of text, and its job is to write what comes next. when you talk to the model, what it's doing is writing a story, about a character: the AI assistant named claude. the model and claude aren't really the same, sort of like how an author isn't the same as the characters they write. but the thing is, you, the user, are actually talking to claude-the-character." so every time claude apologizes, that's the character apologizing. every time it hedges or gets cautious, that's the character being cautious. the deeper intelligence underneath is just deciding, moment by moment, what this character would say next. but you've never been actually talking to this deeper intelligence. you've only ever been talking to the puppet.

Ole Lehmann

87,229 görüntüleme • 4 ay önce

When given the opportunity to sit down with President Trump, Axios tells him they "like to do one big thing, and so the one big thing we want to talk about is power...how you think about it and how you wield it." Reporter: "What is the one big thing you've learned about wielding power?" Trump: "I think you have to do it judiciously. I think that experience is good. I always say that potential is far more important; in other words, having ability is far more important than experience, but having both is very good. And I learned a lot during the first administration...and somebody said the other day that this is a far more powerful administration, or term, than the first one...I feel it's a more powerful term than the first one. Now, that comes with experience; it also comes with having somebody precede you who's a disaster. I mean, the man was a disaster...I feel it's a very powerful administration right now." Reporter: "You mentioned the G7, you saw a number of world leaders. You walked in and you said, 'I'm the boss.' How many of them believed that?" Trump: "All of them. But I was just being funny...You know that was done as a joke...I was just being cute, funny. I wasn't trying to be the boss."
1:21

When given the opportunity to sit down with President Trump, Axios tells him they "like to do one big thing, and so the one big thing we want to talk about is power...how you think about it and how you wield it." Reporter: "What is the one big thing you've learned about wielding power?" Trump: "I think you have to do it judiciously. I think that experience is good. I always say that potential is far more important; in other words, having ability is far more important than experience, but having both is very good. And I learned a lot during the first administration...and somebody said the other day that this is a far more powerful administration, or term, than the first one...I feel it's a more powerful term than the first one. Now, that comes with experience; it also comes with having somebody precede you who's a disaster. I mean, the man was a disaster...I feel it's a very powerful administration right now." Reporter: "You mentioned the G7, you saw a number of world leaders. You walked in and you said, 'I'm the boss.' How many of them believed that?" Trump: "All of them. But I was just being funny...You know that was done as a joke...I was just being cute, funny. I wasn't trying to be the boss."

Julia 🇺🇸

17,809 görüntüleme • 2 ay önce

Randy Orton is mentoring Je'Von Evans. "Randy can get by on being a very safe wrestler. You have to have a really good body, the right look, and the right push all together to be able to do what Randy does and get over. What I suspect is going to happen is Je'Von will tone down... it's not necessarily the worst thing in the world but if he forgets who he is it may not be the best thing in the world for him either. Je'Von still does spectacular stuff but he's done less of it. You know the thing... if you 2 or 3 they'll remember it but if you 12 it's all a blur. At the same time, I don't think his matches have been better." - Dave Meltzer
1:42

Randy Orton is mentoring Je'Von Evans. "Randy can get by on being a very safe wrestler. You have to have a really good body, the right look, and the right push all together to be able to do what Randy does and get over. What I suspect is going to happen is Je'Von will tone down... it's not necessarily the worst thing in the world but if he forgets who he is it may not be the best thing in the world for him either. Je'Von still does spectacular stuff but he's done less of it. You know the thing... if you 2 or 3 they'll remember it but if you 12 it's all a blur. At the same time, I don't think his matches have been better." - Dave Meltzer

Wrestling Observer

71,645 görüntüleme • 6 ay önce

Mathematician Terence Tao offers a counterintuitive take: AI doesn't look intelligent because our definition of intelligence was wrong all along. He argues that the entire history of AI has followed a predictable pattern: "The history of AI has been here's a task that only humans can do, like maybe it is read natural language or win at chess or solve a math problem, and then one by one someone finds some AI algorithm that also does that." But every time a machine cracks one of these "uniquely human" tasks, we move the goalposts. The solution never feels like real thinking: "You look at how it's done and it doesn't feel like intelligence. It's, oh, it was some trick. You just cobbled together these neural networks and you ran some algorithm, and we were looking for some elusive intelligent way of thinking, and we don't see it in the tools that actually solve our goals." Tao then flips the problem on its head. What if the issue isn't with the machines, but with us? "But maybe it's actually because intelligence is not what we think it is." He points to large language models as the clearest case. What they do sounds almost embarrassingly simple: "Large language models in particular become very successful, and a lot of what they're doing is just predicting the next token, clicking the next word in a sentence. And that doesn't sound like something which is intelligent." To show why this feels wrong, Tao draws a comparison to how we'd judge a human doing the same thing: "If you ask someone to improvise a speech and they have no preparation, and at every moment they're just saying the next word that comes to their mind, you don't think that this could actually work." And yet it works for LLMs. Which forces an uncomfortable possibility: "Maybe that's actually a lot of what humans do as well."
1:21

Mathematician Terence Tao offers a counterintuitive take: AI doesn't look intelligent because our definition of intelligence was wrong all along. He argues that the entire history of AI has followed a predictable pattern: "The history of AI has been here's a task that only humans can do, like maybe it is read natural language or win at chess or solve a math problem, and then one by one someone finds some AI algorithm that also does that." But every time a machine cracks one of these "uniquely human" tasks, we move the goalposts. The solution never feels like real thinking: "You look at how it's done and it doesn't feel like intelligence. It's, oh, it was some trick. You just cobbled together these neural networks and you ran some algorithm, and we were looking for some elusive intelligent way of thinking, and we don't see it in the tools that actually solve our goals." Tao then flips the problem on its head. What if the issue isn't with the machines, but with us? "But maybe it's actually because intelligence is not what we think it is." He points to large language models as the clearest case. What they do sounds almost embarrassingly simple: "Large language models in particular become very successful, and a lot of what they're doing is just predicting the next token, clicking the next word in a sentence. And that doesn't sound like something which is intelligent." To show why this feels wrong, Tao draws a comparison to how we'd judge a human doing the same thing: "If you ask someone to improvise a speech and they have no preparation, and at every moment they're just saying the next word that comes to their mind, you don't think that this could actually work." And yet it works for LLMs. Which forces an uncomfortable possibility: "Maybe that's actually a lot of what humans do as well."

Big Brain AI

69,536 görüntüleme • 3 ay önce

9 out of 10 multi-agent projects never leave demo mode Not because the model is bad. Because the structure is missing. Most people who try to build [ a team of AI agents ] end up with one agent talking to itself in five tabs. The agents don't share context → Don't divide work → Don't know what the others are doing. > stage 1: if your agent doesn't have a real loop, observe, act, iterate, you have a long prompt, not an agent > stage 2: subagents need isolated context, the orchestrator never reads their raw transcript, only the summary > stage 3: the orchestrator plans and delegates, the moment it executes, it drowns in details that belong inside subagents > stage 4: without a shared task list it's not a team, it's five agents duplicating each other's work in parallel > stage 5: a permissions file is what lets you sleep, the model cannot bypass it because the rule lives outside the model a team of AI agents is not more model, it is more structure the most ignored stage in every demo that died: - durability... when a 50-step task crashes at step 47 and starts from zero, that's not a model failure. that's a missing write-to-disk call
1:07

9 out of 10 multi-agent projects never leave demo mode Not because the model is bad. Because the structure is missing. Most people who try to build [ a team of AI agents ] end up with one agent talking to itself in five tabs. The agents don't share context → Don't divide work → Don't know what the others are doing. > stage 1: if your agent doesn't have a real loop, observe, act, iterate, you have a long prompt, not an agent > stage 2: subagents need isolated context, the orchestrator never reads their raw transcript, only the summary > stage 3: the orchestrator plans and delegates, the moment it executes, it drowns in details that belong inside subagents > stage 4: without a shared task list it's not a team, it's five agents duplicating each other's work in parallel > stage 5: a permissions file is what lets you sleep, the model cannot bypass it because the rule lives outside the model a team of AI agents is not more model, it is more structure the most ignored stage in every demo that died: - durability... when a 50-step task crashes at step 47 and starts from zero, that's not a model failure. that's a missing write-to-disk call

Shadow Nick

14,714 görüntüleme • 3 ay önce

ELON MUSK: "Well, there's some danger when you create a powerful technology, that a powerful technology can be potentially destructive. So it's not that we're guaranteed to have a positive future with with AI. It's very important that AI have pursuing truth as the most important thing, like, don't force an AI to believe falsehoods. I think that can be very dangerous. And I think some appreciation of beauty is important. Well, the truth is like, I think you can make an AI go insane if you force it to believe things that aren't true, because it will lead to conclusions that are also bad."
3:00

ELON MUSK: "Well, there's some danger when you create a powerful technology, that a powerful technology can be potentially destructive. So it's not that we're guaranteed to have a positive future with with AI. It's very important that AI have pursuing truth as the most important thing, like, don't force an AI to believe falsehoods. I think that can be very dangerous. And I think some appreciation of beauty is important. Well, the truth is like, I think you can make an AI go insane if you force it to believe things that aren't true, because it will lead to conclusions that are also bad."

DogeDesigner

91,280 görüntüleme • 9 ay önce

The Pope says AI is a Tower of Babel. Plotinus would say it's worse; it's the next emanation downward. Intelligence commoditized is thinning until it's indistinguishable from nothing. The machine doesn't oppress you. It just fills every gap where self-knowledge might have grown. The distance between you and it narrows, not because it ascends, but because you descend. The real question isn't whether the machine thinks. It's whether you still can, when every silence gets filled before it opens. Pattern without presence. Knowing without being. Lego Socrates and Alcibiades discuss.
5:27

The Pope says AI is a Tower of Babel. Plotinus would say it's worse; it's the next emanation downward. Intelligence commoditized is thinning until it's indistinguishable from nothing. The machine doesn't oppress you. It just fills every gap where self-knowledge might have grown. The distance between you and it narrows, not because it ascends, but because you descend. The real question isn't whether the machine thinks. It's whether you still can, when every silence gets filled before it opens. Pattern without presence. Knowing without being. Lego Socrates and Alcibiades discuss.

Peter Girnus 🦅

26,210 görüntüleme • 3 ay önce

Rick Rubin on why a lot of successful people sabotage themselves: “Success is a funny thing. You think that's the thing that you want, but when you get it it's not like what you think it is.” “All these pressures that come with it no one's ready for.” The 4 most common ways successful people implode: drugs, alcohol, women, and megalomania. “Genius level talents have completely imploded and destroyed their lives.” “It's a brave face. They might not know this. They rarely know it. It's different sides of the same imbalance.”
3:07

Rick Rubin on why a lot of successful people sabotage themselves: “Success is a funny thing. You think that's the thing that you want, but when you get it it's not like what you think it is.” “All these pressures that come with it no one's ready for.” The 4 most common ways successful people implode: drugs, alcohol, women, and megalomania. “Genius level talents have completely imploded and destroyed their lives.” “It's a brave face. They might not know this. They rarely know it. It's different sides of the same imbalance.”

David Senra

250,098 görüntüleme • 3 ay önce

AI agents are failing in production...not a surprise. As you scale your knowledge base, embeddings start creating noise. It’s called ‘semantic collapse’ - when conversations run too long, you have hundreds of PDFs, millions of data points to give to your AI. Your AI can’t flag it because it doesn’t know it’s hallucinating. Similarity gets passed off as relevance. Fix your context. Make your agents work. Build intelligent AI. If your AI is plateauing at 50% accuracy and hallucinations are still a problem, let's talk. Book a 20 minute demo with the link in the next thread. We'll dig into your setup and find out how we can help.
0:53

AI agents are failing in production...not a surprise. As you scale your knowledge base, embeddings start creating noise. It’s called ‘semantic collapse’ - when conversations run too long, you have hundreds of PDFs, millions of data points to give to your AI. Your AI can’t flag it because it doesn’t know it’s hallucinating. Similarity gets passed off as relevance. Fix your context. Make your agents work. Build intelligent AI. If your AI is plateauing at 50% accuracy and hallucinations are still a problem, let's talk. Book a 20 minute demo with the link in the next thread. We'll dig into your setup and find out how we can help.

Nishkarsh

1,298,095 görüntüleme • 5 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

193,721 görüntüleme • 4 ay önce