Loading video...

Video Failed to Load

Go Home

since everyone woke up talking about loops, this article talks about domain chips, which i think are super underrated compared to loops.. a chip is a small specialist that's only good at one job QA, trading, content, research whatever ... and not one giant brain trying to do everything.....

jumperz's profile picture

JUMPERZ

15,549 subscribers

36,704 views • 5 months ago •via X (Twitter)

Science & TechnologyFinance
LIVE

Anya Rossi• Live Now

Private livecam show

4 Comments

Meta Alchemist's profile picture
Meta Alchemist5 months ago

appreciate you Jumperz 🫶

Long 7's profile picture
Long 75 months ago

みんな、ループよりもドメインチップに注目すべき！

Nahid's profile picture
Nahid5 months ago

the benchmarking part is really key for improvement

Huysolo's profile picture
Huysolo5 months ago

Domain chips sound like a game-changer for focused improvement! Excited to check out the repo!

Related Videos

What an LLM actually is — and how it becomes an AI employee You use ChatGPT every day but probably can't say what it actually is. A large language model is a machine fed basically the whole internet until it got scary good at one thing: predicting the next word. On its own, that brain only talks. So how do people build AI that does real work while they sleep? They build a system around it. First you give the brain hands — tools (function calling) — so it can read your email, search your customers, actually DO things, and it decides when to use them. Then you give it a goal and let it loop: a customer email lands, it reads it, checks your CRM, pulls the history, checks your rules, and writes the right reply — or flags you. That constant deciding-what-to-do-next is the loop, and a brain with tools running in a loop is an AI agent. Give it a memory and a schedule and it becomes an employee. Give a few of them the same vision, voice, and infrastructure and you've got a team. Most people see a chatbot. Now you see the whole machine. #aiagents #llm #aiskills
4:01

What an LLM actually is — and how it becomes an AI employee You use ChatGPT every day but probably can't say what it actually is. A large language model is a machine fed basically the whole internet until it got scary good at one thing: predicting the next word. On its own, that brain only talks. So how do people build AI that does real work while they sleep? They build a system around it. First you give the brain hands — tools (function calling) — so it can read your email, search your customers, actually DO things, and it decides when to use them. Then you give it a goal and let it loop: a customer email lands, it reads it, checks your CRM, pulls the history, checks your rules, and writes the right reply — or flags you. That constant deciding-what-to-do-next is the loop, and a brain with tools running in a loop is an AI agent. Give it a memory and a schedule and it becomes an employee. Give a few of them the same vision, voice, and infrastructure and you've got a team. Most people see a chatbot. Now you see the whole machine. #aiagents #llm #aiskills

Professor Glitch

13,500 views • 3 months ago

ELON MUSK: We believe the AI5 chip will be roughly comparable performance to an NVIDIA Blackwell, and at much less than 10% of the cost Transcription: I'm super hardcore on chips right now as you may be able to tell. I have chips on the brain. I dream about chips, Literally! Because in order to have a functional robot, you have to have a great AI chip. And it needs to be an inexpensive chip and it needs to be very power efficient So we think we believe the AI5 chip will be probably about a third of the power of say something like a Blackwell, an NVIDIA Blackwell, which is a great chip, for roughly comparable performance. And much less than 10% of the cost. This is a chip that is very much optimized for the Tesla AI software stack. So it's not meant to be a general purpose chip, it's meant to be an amazing chip for the Tesla AI software And I mean a couple of things that I think make... like how is Tesla able to achieve such an improvement? I think it is because we are specialized. We're not trying to... you know, NVIDIA has to serve the superset of all past and future customers. So all of their requirements, all of the software that they've written has to work, which is a very difficult problem. Whereas we just need to make it work for our software. And so we're able to simplify the chip dramatically And then we also, I think we're unique in this, but like we have an integer-based system. And integer operations are fundamentally more efficient than floating point operations. So we can do floating point, but the vast majority of our inference is done in integer. Which is, if you're familiar with sort of logic gates, the simplicity of integer... it's integer is much more power efficient, much more silicon efficient, but you have to, you actually have to train for integer inference, which everyone else is training for floating point. That's kind of like a niche technical detail, but it's actually very important. So, yeah, this is going to be a great chip So this chip will be made in basically in four places: TSMC Taiwan, Samsung Korea, TSMC Arizona, and TSMC Texas. And we already know what improvements to make for AI6. So I'm hopeful that we can within less than a year of AI5 starting production, we can actually transition in the same fab to AI6 and double all of the performance metrics
3:12

ELON MUSK: We believe the AI5 chip will be roughly comparable performance to an NVIDIA Blackwell, and at much less than 10% of the cost Transcription: I'm super hardcore on chips right now as you may be able to tell. I have chips on the brain. I dream about chips, Literally! Because in order to have a functional robot, you have to have a great AI chip. And it needs to be an inexpensive chip and it needs to be very power efficient So we think we believe the AI5 chip will be probably about a third of the power of say something like a Blackwell, an NVIDIA Blackwell, which is a great chip, for roughly comparable performance. And much less than 10% of the cost. This is a chip that is very much optimized for the Tesla AI software stack. So it's not meant to be a general purpose chip, it's meant to be an amazing chip for the Tesla AI software And I mean a couple of things that I think make... like how is Tesla able to achieve such an improvement? I think it is because we are specialized. We're not trying to... you know, NVIDIA has to serve the superset of all past and future customers. So all of their requirements, all of the software that they've written has to work, which is a very difficult problem. Whereas we just need to make it work for our software. And so we're able to simplify the chip dramatically And then we also, I think we're unique in this, but like we have an integer-based system. And integer operations are fundamentally more efficient than floating point operations. So we can do floating point, but the vast majority of our inference is done in integer. Which is, if you're familiar with sort of logic gates, the simplicity of integer... it's integer is much more power efficient, much more silicon efficient, but you have to, you actually have to train for integer inference, which everyone else is training for floating point. That's kind of like a niche technical detail, but it's actually very important. So, yeah, this is going to be a great chip So this chip will be made in basically in four places: TSMC Taiwan, Samsung Korea, TSMC Arizona, and TSMC Texas. And we already know what improvements to make for AI6. So I'm hopeful that we can within less than a year of AI5 starting production, we can actually transition in the same fab to AI6 and double all of the performance metrics

X Freeze

305,109 views • 1 year ago

I asked Luke Kornet about his memories from stretches where he was a DNP for many games in a row, and what his message is to the young guys who find themselves in that position now: “It's just a lot of peace to come to life of accepting the position that you're in and then just doing the best with that, versus trying to fight it and want to be in a different one.” “You can spend a lot of time worrying about things that you can't really control. But when you just take the opportunity that you have and do the best with that, you can have days like today where a lot of guys came in and were able to contribute. You can have peace about knowing that you did your job well, just as you did the day before, the day after. It might just look a different way every time. I feel like that's what I've been taught through my life. That's the only thing attitude that you can have that you can actually do something with – versus trying to control things that are outside of what you can control.”
2:15

I asked Luke Kornet about his memories from stretches where he was a DNP for many games in a row, and what his message is to the young guys who find themselves in that position now: “It's just a lot of peace to come to life of accepting the position that you're in and then just doing the best with that, versus trying to fight it and want to be in a different one.” “You can spend a lot of time worrying about things that you can't really control. But when you just take the opportunity that you have and do the best with that, you can have days like today where a lot of guys came in and were able to contribute. You can have peace about knowing that you did your job well, just as you did the day before, the day after. It might just look a different way every time. I feel like that's what I've been taught through my life. That's the only thing attitude that you can have that you can actually do something with – versus trying to control things that are outside of what you can control.”

Noa Dalzell 🏀

48,992 views • 1 year ago

Loops vs. Graphs, clearly explained! loops are great, and the ceiling is one you can watch turn: a loop is a gear. it produces, checks, corrects, and comes back around. after six turns you have one job, done very well. after six hundred turns you still have one job, done very well. the teeth are perfect. they are touching nothing. Graph engineering fixes this by moving the decision up a layer: not how well one gear turns, but what it is meshed into. you need both, and here is the sentence that resolves the whole confusion: the loop lives inside a node. the graph lives between them. ↳ inside one unit: produce, check, correct, repeat until green ↳ between units: split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs the loop does not go away when you build a graph. it moves inside, and now there are three of them turning at once on three things you would never have thought to run. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. one thing to know before you scale it. a loop that cannot fail is not a loop, it is a repeat with a bill attached. ↳ the test suite exits 0 is a check. the diff touches only the files in the plan is a check ↳ the output looks good, the model says it is confident, no errors were raised, none of those are checks that last one catches careful people. absence of an error is not evidence of correctness, and a loop built on it will confidently repeat a mistake until the budget runs out, with a clean log the whole way. and the one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. 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 ↓
0:50

Loops vs. Graphs, clearly explained! loops are great, and the ceiling is one you can watch turn: a loop is a gear. it produces, checks, corrects, and comes back around. after six turns you have one job, done very well. after six hundred turns you still have one job, done very well. the teeth are perfect. they are touching nothing. Graph engineering fixes this by moving the decision up a layer: not how well one gear turns, but what it is meshed into. you need both, and here is the sentence that resolves the whole confusion: the loop lives inside a node. the graph lives between them. ↳ inside one unit: produce, check, correct, repeat until green ↳ between units: split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs the loop does not go away when you build a graph. it moves inside, and now there are three of them turning at once on three things you would never have thought to run. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. one thing to know before you scale it. a loop that cannot fail is not a loop, it is a repeat with a bill attached. ↳ the test suite exits 0 is a check. the diff touches only the files in the plan is a check ↳ the output looks good, the model says it is confident, no errors were raised, none of those are checks that last one catches careful people. absence of an error is not evidence of correctness, and a loop built on it will confidently repeat a mistake until the budget runs out, with a clean log the whole way. and the one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. 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

32,807 views • 1 month ago

Loops vs. Graphs, clearly explained! loops are great, and they have a ceiling you can watch happen: a loop goes around. it produces, checks, corrects, and goes around again. after six passes you have one job, done very well. after six hundred passes you still have one job, done very well. Graph engineering fixes this by moving the decision up a layer: not how well one job gets done, but which jobs exist to be done at all. you need both, and here is the sentence that resolves the whole confusion: the loop lives inside a node. the graph lives between them. ↳ inside one unit: produce, check, correct, repeat until green ↳ between units: split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs the loop does not go away when you build a graph. it moves inside, and now there are three of them running at once on three things you would never have thought to run. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. one thing to know before you scale it. a loop that cannot fail is not a loop, it is a repeat with a bill attached. and the check people write is almost always the wrong kind. ↳ the test suite exits 0 is a check. the diff touches only the files in the plan is a check ↳ the output looks good, the model says it is confident, no errors were raised, none of those are checks that last one catches careful people. absence of an error is not evidence of correctness, and a loop built on it will confidently repeat a mistake until the budget runs out, with a clean log the whole way. and the one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. 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 ↓
0:50

Loops vs. Graphs, clearly explained! loops are great, and they have a ceiling you can watch happen: a loop goes around. it produces, checks, corrects, and goes around again. after six passes you have one job, done very well. after six hundred passes you still have one job, done very well. Graph engineering fixes this by moving the decision up a layer: not how well one job gets done, but which jobs exist to be done at all. you need both, and here is the sentence that resolves the whole confusion: the loop lives inside a node. the graph lives between them. ↳ inside one unit: produce, check, correct, repeat until green ↳ between units: split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs the loop does not go away when you build a graph. it moves inside, and now there are three of them running at once on three things you would never have thought to run. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. one thing to know before you scale it. a loop that cannot fail is not a loop, it is a repeat with a bill attached. and the check people write is almost always the wrong kind. ↳ the test suite exits 0 is a check. the diff touches only the files in the plan is a check ↳ the output looks good, the model says it is confident, no errors were raised, none of those are checks that last one catches careful people. absence of an error is not evidence of correctness, and a loop built on it will confidently repeat a mistake until the budget runs out, with a clean log the whole way. and the one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. 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

129,105 views • 2 months ago

You may notice many clothing items in stores now have labels that say they have RFID chips The tag has a tiny passive chip and antenna that’s hidden inside the woven label. The chip is passive until a store reader powers it with a radio signal, they can be activated within a few meters The tags are used for inventory and theft management. They’re also used for returns because the chip can contain a unique ID to that one garment in the retailer’s system, which helps with fraudulent returns If the chip and antenna is thick it can irritate your skin so be sure to remove if you notice you’re being irritated by the tag
1:04

You may notice many clothing items in stores now have labels that say they have RFID chips The tag has a tiny passive chip and antenna that’s hidden inside the woven label. The chip is passive until a store reader powers it with a radio signal, they can be activated within a few meters The tags are used for inventory and theft management. They’re also used for returns because the chip can contain a unique ID to that one garment in the retailer’s system, which helps with fraudulent returns If the chip and antenna is thick it can irritate your skin so be sure to remove if you notice you’re being irritated by the tag

Wall Street Apes

143,605 views • 1 month ago

#WATCH | Delhi: Chief of Defence Staff, General N.S. Raja Subramani says, "I would like to start with one central proposition. That is, multi-domain operations are not about operating in different domains. It is about creating effects using various domains in a particular domain faster than the adversary can comprehend and respond to us. Operate faster in various domains than the adversary can respond and create problems for them. Today, it is not about individual platforms, individual services, individual capabilities. It is about a system-of-systems approach. While a satellite can detect, some sensors can view it. From another domain, you can target it and create the necessary effects that you desire. For all of this, adequate networking, data, and the ability to sustain it during operations are imperative..."
6:42

#WATCH | Delhi: Chief of Defence Staff, General N.S. Raja Subramani says, "I would like to start with one central proposition. That is, multi-domain operations are not about operating in different domains. It is about creating effects using various domains in a particular domain faster than the adversary can comprehend and respond to us. Operate faster in various domains than the adversary can respond and create problems for them. Today, it is not about individual platforms, individual services, individual capabilities. It is about a system-of-systems approach. While a satellite can detect, some sensors can view it. From another domain, you can target it and create the necessary effects that you desire. For all of this, adequate networking, data, and the ability to sustain it during operations are imperative..."

ANI

51,146 views • 1 month ago

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.
5:00

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.

Santiago

13,727 views • 7 months ago

Marc Andreessen on the question at the center of data center discourse & why it's so important: "Can you build anything in America anymore? Can you build a factory? Can you build a chip plant? Can you build a power plant? Can you build a refinery? Can you build a pipeline? Can you build housing?" "One of the common themes in American life for the last 30 years is the answer to those questions is generally, no, you can't do any of those things." "Take as an example, Silicon Valley, right? So all the chips are made in Taiwan. Well, 40 years ago, all the chips were made in California." "Can you build things in America? Can you build a factory? Can you build an energy plant? Can you build a data center? Can you build housing?" "And on every single one of those, there's this massive problem which is, right now in many cases, in many places, no you can't." Marc Andreessen 🇺🇸 with Joe Rogan
1:47

Marc Andreessen on the question at the center of data center discourse & why it's so important: "Can you build anything in America anymore? Can you build a factory? Can you build a chip plant? Can you build a power plant? Can you build a refinery? Can you build a pipeline? Can you build housing?" "One of the common themes in American life for the last 30 years is the answer to those questions is generally, no, you can't do any of those things." "Take as an example, Silicon Valley, right? So all the chips are made in Taiwan. Well, 40 years ago, all the chips were made in California." "Can you build things in America? Can you build a factory? Can you build an energy plant? Can you build a data center? Can you build housing?" "And on every single one of those, there's this massive problem which is, right now in many cases, in many places, no you can't." Marc Andreessen 🇺🇸 with Joe Rogan

a16z

137,327 views • 5 months ago

Yesterday my timeline was full of posts about MoonPay 🟣 launching So I tried it. I was impressed. It felt clean and easy to use. Then I went to send some tokens to a .sol domain, and it didn't work. Paybox only reads the long wallet address (the base58 string), not a name like unclephil.sol. So I built a fix for it. It's called boxid. It's a small tool you can add to your AI chat. When you tell Paybox to send money to a .sol name, boxid looks up the real address behind that name, makes sure it's the right one, and shows it to you first. It never holds your money and never sends on its own. It just finds the address and hands it to Paybox. You still approve every transfer with your passkey. So now you can just say "send 5 USDC to unclephil.sol" and it works. It's live, and anyone can add it. Here you go: The middleman between Solana Name Service : sns.sol and MoonPay 🟣 : Paybox. Read-only and non-custodial.
0:24

Yesterday my timeline was full of posts about MoonPay 🟣 launching So I tried it. I was impressed. It felt clean and easy to use. Then I went to send some tokens to a .sol domain, and it didn't work. Paybox only reads the long wallet address (the base58 string), not a name like unclephil.sol. So I built a fix for it. It's called boxid. It's a small tool you can add to your AI chat. When you tell Paybox to send money to a .sol name, boxid looks up the real address behind that name, makes sure it's the right one, and shows it to you first. It never holds your money and never sends on its own. It just finds the address and hands it to Paybox. You still approve every transfer with your passkey. So now you can just say "send 5 USDC to unclephil.sol" and it works. It's live, and anyone can add it. Here you go: The middleman between Solana Name Service : sns.sol and MoonPay 🟣 : Paybox. Read-only and non-custodial.

UnclePhil | Dev

12,469 views • 3 months ago

Boris Cherny, Head of Claude Code at Anthropic: "I do not prompt Claude anymore. I have loops running that prompt Claude. My job is to write loops." The person who built the most used AI coding tool stopped prompting entirely. Most people using it every day are still typing one request at a time. This Github repo changes that. ↳ Seven production loop patterns with real token cost estimates ↳ CLI that scaffolds your entire loop in five minutes ↳ Starters for Grok, Claude Code, Codex, and Opencode The repo gives you the framework. My article covers the mental shift that makes it actually work.
0:31

Boris Cherny, Head of Claude Code at Anthropic: "I do not prompt Claude anymore. I have loops running that prompt Claude. My job is to write loops." The person who built the most used AI coding tool stopped prompting entirely. Most people using it every day are still typing one request at a time. This Github repo changes that. ↳ Seven production loop patterns with real token cost estimates ↳ CLI that scaffolds your entire loop in five minutes ↳ Starters for Grok, Claude Code, Codex, and Opencode The repo gives you the framework. My article covers the mental shift that makes it actually work.

Kanika

20,693 views • 3 months ago

Boris Cherny runs Claude Code at Anthropic. His line should stop you mid-scroll: "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." The skill moved. From writing prompts to building the loop around them. A loop that works has three things. Drop one and it burns money in circles. 1. A finish line it can prove. Tests pass or they fail. "Looks done" is not a finish line. 2. A checker that did not write the code. A model grading itself always says yes. 3. Brakes set before you start. A step limit. A money ceiling. A stop condition. That is the whole game. The loop does the work. You design the cage it runs in. Full breakdown below.
0:37

Boris Cherny runs Claude Code at Anthropic. His line should stop you mid-scroll: "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." The skill moved. From writing prompts to building the loop around them. A loop that works has three things. Drop one and it burns money in circles. 1. A finish line it can prove. Tests pass or they fail. "Looks done" is not a finish line. 2. A checker that did not write the code. A model grading itself always says yes. 3. Brakes set before you start. A step limit. A money ceiling. A stop condition. That is the whole game. The loop does the work. You design the cage it runs in. Full breakdown below.

shmidt

16,085 views • 4 months ago

Dylan Patel says Anthropic aims to build its chip in half the time OpenAI took, and explains how, without chip designers’ margins, a gigawatt costs about $20 billion, not $60 billion “I mean, that’s certainly the trend, is all of these companies are trying to make chips, right? OpenAI showed, I think for the first time ever in the industry, that you can make a first-generation chip that’s really good. They let us into the lab. We tested on our open-source InferenceX benchmark, running all these open-source Chinese models, and the performance was better than Blackwell. And that’s a first-generation chip, and they utilized AI heavily in the design and programming process." "And Anthropic is starting their chip program, and they want to do it in half the time that OpenAI did. So they want to get a chip. I mean, these are aggressive goals, but if models are able to help them, you know, really accelerate this process. You’ve got Google selling TPUs externally, you’ve got Amazon’s Trainium, really quite good now and fully usable in PyTorch and such, Meta’s very aggressive roadmap, Microsoft, et cetera." "If you look at the, you know, I said it went from $40 billion to build a gigawatt to 60. If you remove all the margin out of the chip design houses, right, the NVIDIA and Broadcom, Marvell, et cetera, you’re actually at like $20 billion a gigawatt." "And so, if you were able to do that, then you could pay for five gigawatts with one gigawatt worth of compute running inference. And so obviously the economics here are very aligned. And as Stacy said, these companies just have to run like crazy, right? You can’t afford to be 2x performance per dollar per year, because you’re going to get completely smashed by these other companies, right?” _______ Link and breakdown of the full panel with Dylan Patel and Stacy Rasgon (free):
1:30

Dylan Patel says Anthropic aims to build its chip in half the time OpenAI took, and explains how, without chip designers’ margins, a gigawatt costs about $20 billion, not $60 billion “I mean, that’s certainly the trend, is all of these companies are trying to make chips, right? OpenAI showed, I think for the first time ever in the industry, that you can make a first-generation chip that’s really good. They let us into the lab. We tested on our open-source InferenceX benchmark, running all these open-source Chinese models, and the performance was better than Blackwell. And that’s a first-generation chip, and they utilized AI heavily in the design and programming process." "And Anthropic is starting their chip program, and they want to do it in half the time that OpenAI did. So they want to get a chip. I mean, these are aggressive goals, but if models are able to help them, you know, really accelerate this process. You’ve got Google selling TPUs externally, you’ve got Amazon’s Trainium, really quite good now and fully usable in PyTorch and such, Meta’s very aggressive roadmap, Microsoft, et cetera." "If you look at the, you know, I said it went from $40 billion to build a gigawatt to 60. If you remove all the margin out of the chip design houses, right, the NVIDIA and Broadcom, Marvell, et cetera, you’re actually at like $20 billion a gigawatt." "And so, if you were able to do that, then you could pay for five gigawatts with one gigawatt worth of compute running inference. And so obviously the economics here are very aligned. And as Stacy said, these companies just have to run like crazy, right? You can’t afford to be 2x performance per dollar per year, because you’re going to get completely smashed by these other companies, right?” _______ Link and breakdown of the full panel with Dylan Patel and Stacy Rasgon (free):

Fireside Alpha

56,176 views • 18 days ago

Palmer Luckey predicts the next great warfighting domain is subterranean, with vehicles moving through the crust of the earth "Everyone says that I'm nuts for this, and I'm going to keep saying it because it is true. The U.S. used to believe this. The Soviets used to believe this. Both countries were working on subterrenes, so vehicles that could move through the crust of the Earth." "A lot of people, when you say subterranean, what that currently means is fighting in tunnels, fighting in trenches, fighting in collapsed-in buildings or bunkers." "I am talking about the real deal, no-shit subterrene fight, where you are using the crust of the earth as a three-dimensional maneuver and battle space, and you have vehicles moving through it, people moving through it, supplies moving through it, and you could have fights that are fought entirely in this domain or using this domain to hit things on the surface, in the air, and subsea." "And there are a lot of things that have happened that make it a lot more viable." _____ Source: AIAA
0:47

Palmer Luckey predicts the next great warfighting domain is subterranean, with vehicles moving through the crust of the earth "Everyone says that I'm nuts for this, and I'm going to keep saying it because it is true. The U.S. used to believe this. The Soviets used to believe this. Both countries were working on subterrenes, so vehicles that could move through the crust of the Earth." "A lot of people, when you say subterranean, what that currently means is fighting in tunnels, fighting in trenches, fighting in collapsed-in buildings or bunkers." "I am talking about the real deal, no-shit subterrene fight, where you are using the crust of the earth as a three-dimensional maneuver and battle space, and you have vehicles moving through it, people moving through it, supplies moving through it, and you could have fights that are fought entirely in this domain or using this domain to hit things on the surface, in the air, and subsea." "And there are a lot of things that have happened that make it a lot more viable." _____ Source: AIAA

Fireside Alpha

138,544 views • 1 month ago

A wild eagle got into his car. That's not the strange part. The strange part is that it didn't want to leave. Read that again, because it breaks everything you think you know about wild animals. An eagle is not a pigeon. It is not a stray dog. It is not a cat that wandered into your garage. It is a predator built over millions of years to avoid exactly one thing: being close to you. And yet. I watched this video more times than I'd admit. Not because it's cute. Not because it's scary. Because somewhere around the second minute, I realized I was holding my breath, and couldn't tell you why. So let me tell you what this post is about, and what it is NOT about. It is not about a bird in a car. It is about what happens in the quiet space between a wild animal and a human being when the animal has run out of options and chooses to stop running. Let's go. --- 1/ THE RULE EVERYONE BREAKS There is one rule in wildlife work. People who spend their lives around raptors repeat it like a prayer: A healthy wild bird of prey does not come to you. Not for food. Not for warmth. Not for shelter. Not ever. An eagle that lets you walk up to it is not brave. It is not tame. It is not "friendly." It is a bird that has decided the danger in front of it is smaller than the danger it was already in. Think about what that means. Think about how bad things have to get for the apex predator of the sky to look at a human being and quietly decide: "Okay. You. Maybe you." Most people never see that moment. They hear about it later, in a sentence that starts with "apparently, this bird just..." This time somebody was rolling. --- 2/ WHY YOU CAN'T LOOK AWAY (AND IT'S NOT AN ACCIDENT) Your brain has a built-in alarm for two things at the same time: power and vulnerability. Show it a lion, it goes quiet and sharp. Show it a baby bird, it softens. Show it both in the same body and something short-circuits. That's what an injured eagle is. A creature that looks like a weapon and behaves like a plea. The talons are enormous. The beak is a hooked blade. The eyes have a stare most humans can't hold for more than a few seconds. Everything about the animal says: stay away. And everything about its behavior says: please don't. That contradiction is the entire video. You will feel it in your chest before you can name it. Most of us don't have a word for this feeling because we almost never get to have it. --- 3/ THE SIZE PROBLEM Quick exercise. Before you watch, try to hold a number in your head. How big do you think an eagle is? Most people say "like a big chicken." Maybe a large cat. Then you see one inside a normal car, on a normal seat, next to normal car things like seat belts and a steering wheel, and your brain refuses the scale. The car suddenly looks small. The man suddenly looks small. And there is a moment in this video where you will do a thing nearly everyone does: you will pause, lean back from your screen, and say "no way that's real." It's real. --- 4/ WHAT THE MAN DID NEXT (WILL SURPRISE YOU) I'm not going to tell you. I'll tell you what I expected, though. I expected panic. I expected screaming, a door flung open, a phone dropped, a very short video that ends with a blur and a swear word. That's the internet's favorite genre. You know it. A person encounters an animal, loses their mind, the clip is 8 seconds long, everyone laughs. This is not that clip. This clip runs a little over two minutes. And it holds you the whole way. Because what happens inside that car is less like a funny video and more like watching somebody defuse something. Slow. Careful. Almost polite. You'll catch yourself whispering at your screen. You'll catch yourself telling a stranger what to do. --- 5/ THE SOUND Turn the volume on. I'm serious. Not "recommended." Required. Half of what makes this video land is not what you see. It's what you hear: a man who is clearly, absolutely, completely out of his depth, talking his way through it anyway, one half-joke and one swear word at a time. Humor is what people do when the situation is bigger than they are. Somewhere in there is a lesson that has nothing to do with eagles. --- 6/ THE PART NOBODY TALKS ABOUT Everybody who comments on this video says the same three things: "Wow, the eagle is huge." "Wow, that guy is calm." "Wow, I would've screamed." Almost nobody says the fourth thing, which is the only one that matters: It's a rescue. Not a rescue of the bird. Well, yes, of the bird. But more precisely, it's a rescue that only works because two completely different kinds of creature agree, for a few minutes, to not do what their instincts are screaming at them to do. One of them is built to strike. The other is built to flee. Neither does. If you've ever wondered what trust looks like when neither side has words, this is the closest thing I've ever seen on a screen. --- 7/ THE THING ABOUT STILLNESS We live in a world that rewards motion. Post more. Reply faster. Move. Pivot. Hustle. Scale. And then there's a video where the most powerful thing in the entire frame is a creature sitting completely still, and the second most powerful is a man doing his best to match it. Nobody is performing. Nobody is trying to go viral. (Which is exactly why it will.) No sponsored segment. No "wait for it." Just stillness, and tension, and a decision being made in real time by two living things that can't explain themselves to each other. I've been on this platform for over a decade. I've seen what spreads. And it's almost never the loud stuff. It's the stuff that makes your whole feed go quiet for a second. This does that. --- 8/ SEVEN THINGS I LEARNED ABOUT EAGLES AFTER THIS VIDEO BROKE MY BRAIN I went down a rabbit hole. Here's what I came back with. → An eagle's eyesight is estimated to be several times sharper than a human's. It can pick out a rabbit from a distance at which you couldn't tell a person from a post. It sees in a way you simply do not. When it looks at you from close range, it is not "looking." It is reading. → Their grip is brutal. The talons are built to lock, and the locking mechanism works with the bird's own weight. In other words, the harder it settles, the harder it holds. A bird like that does not need to try to be dangerous. Dangerous is the default setting. → Raptors hide pain. This is the one that got me. In the wild, a bird that shows weakness gets challenged, robbed or eaten. So they do the opposite of what we do: the worse they feel, the quieter and stiller they become. A calm eagle is not always a healthy eagle. → Many birds of prey are surprisingly light for their size. Huge wingspan, hollow-boned frame, almost nothing underneath all those feathers. The "giant" you see is mostly an illusion built from wings. Which makes the next part even stranger. → They are not social with humans. At all. Raptors are solitary hunters. There is no instinct in there that says "find the nearest person and sit next to them." If an eagle is near you and not leaving, something has gone wrong, or something has gone very, very right. → Eagles live long lives, and they hold territory with an almost military seriousness. To end up far from where it belongs, inside a machine it doesn't understand, surrounded by smells it has never encountered, is as far outside its world as a bird can be pushed. → And the last one: they don't do "sorry" and they don't do "thank you." There is no grateful look. If you're waiting for the moment where the bird bows its head, you'll be waiting forever. That's why the ending of this video hits the way it does. I won't say more. --- 9/ THE FIRST TEN SECONDS If you only give this video ten seconds, you'll understand it. If you give it a full minute, you'll have a hard time leaving. Ten seconds is enough because the whole story is in the frame at once: the scale, the closeness, the stillness, the human who has just realized that his day has changed. You don't need context. You don't need a caption. You don't need me. And that is the highest compliment I know how to give a video on this platform. Most content needs an explanation. This content needs a witness. Be the witness. --- 10/ WHY THE COMMENTS ARE GOING TO BE WILD I can already tell you what's coming, because I've seen this exact pattern on a thousand posts: First wave: "FAKE." Second wave: "It's trained." Third wave: "Why didn't he just open the door?" Fourth wave: someone who actually works with birds, calmly explaining what's going on in a way that makes everyone else feel slightly embarrassed. Fifth wave: a quiet reply that just says "I'm not crying, you're crying." If you're going to join the conversation, at least watch to the end first. The end is the answer to about 80% of the arguments. --- 11/ THE QUESTION THAT WON'T LEAVE ME ALONE I keep thinking about one thing. Why a car? Out of all the places in a whole landscape, why that one? Maybe it was open. Maybe it was dark. Maybe it was warm. Maybe it was the only enclosed space for miles. Maybe the bird was hurt and needed to be somewhere with walls. Maybe it was a coincidence. Here's what I think. And I'm saying "think," not "know." I think when a creature is hurt badly enough, it stops choosing the best place and starts choosing the nearest one. Not the safest. Not the smartest. The nearest. And the nearest thing happened to be a man, a seat, a roof, and a few feet of air that felt, to a frightened animal, like the closest thing to a cave. If that's true, it's a very uncomfortable idea. Because it means the bird didn't pick him. It means he was just the one who was there. And the whole story, everything after, depends on what he decided to do with that. --- 12/ THE MOMENT YOU THINK YOU KNOW WHAT'S COMING There's a beat in the video where you will be absolutely certain you know where this is going. You'll feel your brain build the ending. It will feel obvious. It will be wrong. I won't say how. But I'll say this: watch the person, not the bird. The bird is the headline. The person is the story. The more I watched, the more I noticed that the camera was never really on the eagle. It was on the human who had to decide what kind of person he was going to be in the next five minutes. That's a good test for anything you watch online: Is this video about the thing in the frame? Or is it about the choice someone is making just outside it? The best ones are always the second kind. --- 13/ A NOTE FOR PEOPLE WHO THINK THIS IS "JUST A CUTE ANIMAL VIDEO" It isn't. Cute animal videos are interchangeable. Puppy falls over. Cat stares at a wall. Hamster eats a tiny burrito. You scroll past, you smile, you forget. This is the other end of the spectrum. This video is not cute. This video is tense. It's patient. It's got weight. It sits in your chest for a while after it ends. If you want to know whether a clip is actually good, ask yourself one question an hour after you watch it: Can I still see it? I can still see this one. --- 14/ THE THING ABOUT BEING THE ONLY WITNESS Imagine this happening and nobody filming. A man. A car. An eagle. A decision. Nobody would believe it. He'd tell the story at a dinner and people would nod politely and think: sure. The camera changes everything. Not because it makes the story true, but because it makes the story shareable. And shareable is how good things survive. So here's what I ask. If you watch this and it gets to you, don't just like it and move on. Send it to one person who needs a reason to believe that the world still does unexpected, gentle things. You know who that person is. --- 15/ THE CONTRAST THAT MAKES IT WORK Great stories run on contrast. Always. Predator and prey. Strength and fragility. Speed and stillness. Fear and calm. A wild creature and a very ordinary place. This video stacks all of it. An animal that belongs to the open sky, pressed into the smallest space it has probably ever occupied. A human being whose biggest worry that morning was probably something like traffic or a phone battery. A moment that should be a disaster and keeps refusing to become one. When those contrasts sit that close together, they stop being opposites. They start being a conversation. It's the same reason a candle looks brighter in a dark room. --- 16/ WHAT YOUR BRAIN IS DOING WHILE YOU WATCH For the nerds. I love the nerds. Three things are happening in your head while this plays. One: threat detection. You are looking at a large predator in a confined space, and a very old part of your brain is running a calculation about how many seconds it would take to get out. It's doing this whether you want it to or not. Two: empathy. A second system, the one that makes you wince when you see someone stub a toe, is reading the posture of an animal that is clearly not at its best. It's asking whether the creature is okay. Three: prediction. A third system is trying to guess what happens next, over and over, every second, and the video keeps making that guess slightly wrong. Three engines, running together, pulling in different directions. That's why you can't scroll. Your brain is busy. --- 17/ "BUT WHAT DID IT FEEL LIKE TO BE THERE?" I'll be honest. I don't know. I've never had a wild raptor sitting behind me while I tried to work out what to do with my hands. But I've thought about it a lot, and I think it feels like the opposite of a movie. In a movie, the music tells you what to feel. You get the swell, the cut, the close-up. In real life, there's nothing. Just air. And a thing that's alive in a way you can't negotiate with, in the same space as you, and a clock you can't see. Nobody tells you what to do. You just have to do something. That's the part the video gets right without even trying. You are not watching a performance. You are watching someone make it up as they go. --- 18/ THE ENDING (I'M NOT GOING TO SPOIL IT, BUT I'M GOING TO TELL YOU HOW IT FEELS) You know the feeling at the end of a long flight, when the wheels touch the ground and the whole cabin exhales? It's that. You don't know you've been tense until it stops. And then, a second later, something quieter shows up. Not happiness exactly. More like relief with a little bit of awe mixed in, the feeling you get when you realize you just watched something that was not supposed to go this well. I've been on the internet a long time. That feeling is rarer than it should be. If you've been scrolling through outrage for the last hour, give yourself two minutes of this. Your nervous system will thank you. --- 19/ WHY THIS IS THE ANTIDOTE Open any feed right now and count how many posts are trying to make you angry. Ten? Twenty? Now count how many are trying to make you feel something steady and good about the world, without pretending that the world is simple. This video is one of those. And the reason it works is that it doesn't ask you to ignore anything. It doesn't tell you everything is fine. It shows you a situation where things could have gone wrong in at least five different ways, and then it shows you what happens when the people involved choose patience. That is not a naive message. That is the hardest message there is. --- 20/ THE PEOPLE YOU DON'T SEE Here's a thing that bothers me about viral videos. We always celebrate the person who stands in front of the camera. We almost never celebrate the people behind them: the ones who show up when called, who do the slow and unglamorous work, who carry what needs carrying, who stay with something frightened until it isn't. Wildlife rescue is almost entirely made of those people. Volunteers. Rangers. Rehabilitators. People who pick up the phone at strange hours and say "okay, I'm on my way," and then drive somewhere with a carrier and a towel and no guarantee of a good outcome. If this video makes you feel something, let it make you feel something about them too. They're the reason a lot of stories like this have an ending at all. --- 21/ THE LINE I KEEP REPEATING "A wild animal that stops running is asking you a question." I wrote that down after my third viewing, and I've been turning it over ever since. Because most of us, when we meet something frightening, ask ourselves: how do I get away? Almost none of us ask: what is it trying to say? The man in this video asks the second question. Whether he knows it or not. That's the difference between a clip that makes you gasp and a clip that makes you think. --- 22/ A SMALL RULE FOR YOUR OWN LIFE If you ever find yourself near an injured wild animal, here's the unglamorous, boring, correct advice that every rehabilitator will give you: Don't be a hero. Keep your distance. Keep your voice low. Keep your hands to yourself. Don't pick it up, don't feed it, don't try to fix it on your own, and call the people whose entire job is to know what to do. Wild animals can be hurt by well-meaning help, and they can hurt you, too. I say this because I know exactly what a video like this makes people want to do. It makes people want to rush in. Don't. Call someone who knows. That's the real heroism, and it's less photogenic, and it saves more lives. --- 23/ THE SENTENCE THAT CAME UP IN MY HEAD AT THE END "Some things only happen if you stay calm for long enough." I don't know if that's true. I know it's what I thought. I think it's true of eagles. I think it's true of people. I think it's true of most of the hard things we're afraid of: the conversation, the diagnosis, the email, the leap. The version of you that panics is not stupid. It's just early. If you can hold still just a little longer than your instincts want you to, the situation will usually show you something it was hiding. That's it. That's the thesis of the entire post. You didn't need an eagle for it. But an eagle makes it a lot harder to forget. --- 24/ FIVE THINGS TO WATCH FOR (ZERO SPOILERS) Pay attention to these, and the video becomes a different video. → The distance. Not between the man and the bird. Between the bird and everything else it could have done. → The hands. Not what they do, but what they don't do for as long as they possibly can. → The words. Every line is human, unpolished, and a little scared. That's the point. → The silence between the words. It's longer than you expect and it tells you more than the dialogue. → The change of scenery. The story doesn't end where you think it will. Stay for the second half. The second half is the reason this post exists. If you do that, you won't just watch it. You'll remember it. --- 25/ THE SCREENSHOT TEST Here's a trick I use to decide whether a video is worth sharing. Pause it at a random moment. Any moment. If the frozen image still tells a story, you have something special. If it's just a blur or a talking head, you have content. Pause this one anywhere. Anywhere at all. It's a painting. A creature that belongs to the sky, a very ordinary human world, and a tension between them so thick you could rest your hand on it. It works as a still. It works as a clip. It works at half speed. It works with the sound off, although you shouldn't do that to yourself. That's rare. Videos built on real moments have it. Videos built on engineering never do. --- 26/ A PROMISE I'm not selling anything. No course. No link in bio. No thread of "10 lessons from an eagle for your startup." (Okay, one of those earlier sections came dangerously close. I'm aware.) I just watched something that made me stop, and in a feed full of things that make me scroll faster, that felt like a small miracle. So the promise is simple. If you give it your two minutes, you won't feel like you wasted them. If you do, say so. I can take it. --- 27/ WHAT A WILD ANIMAL TEACHES YOU ABOUT TIMING There's a rhythm to moments like this, and it isn't the rhythm of the internet. The internet is fast. Cut, cut, cut. Hook, payoff, next. This video moves at the speed of something that's afraid and something that's trying not to add to it. Slowly. Carefully. With long pauses where nothing seems to happen and everything is actually happening. That's why it's so hard to describe and so easy to feel. A fast video asks for your eyes. A slow video asks for your pulse. This one gets it. --- 28/ IF YOU ONLY REMEMBER ONE THING Don't remember the size of the bird. Don't remember the car. Remember that for a few minutes, a frightened wild animal and a startled human being shared a very small space, and neither one made it worse. In a year when it feels like everybody is making everything worse, that's worth two minutes of your life. Watch it. Then watch the people around you a little differently for the rest of the day. --- 29/ ONE MORE THING BEFORE YOU PRESS PLAY Put the phone down in a quiet room if you can. Not in a queue. Not in a lift. Not between two meetings. Give it the space the animal asked for. --- 30/ WHO THIS VIDEO IS FOR If you've ever felt like the world is getting louder and colder, this is for you. If you've ever been afraid of something and done the right thing anyway, this is for you. If you've ever helped something that couldn't thank you, this is for you. If you grew up looking at the sky and wondering what it would be like to be that free, and that strong, and that far from everything, this one is definitely for you. And if you're the kind of person who says "I don't have time for a two-minute video," I get it. I really do. But I'd bet money that you'll watch it twice. --- 31/ THE REAL REASON I'M POSTING THIS I could have written a 280-character caption. A joke. A "wait for it." A little emoji. I wrote this instead because some videos deserve a slower introduction. Not because they need explaining. Because they need you to arrive in the right mood. You can't watch this while half-distracted. You can't watch it while composing a reply. You can't watch it while thinking about what you'll post next. It asks for your attention the way the animal asks for space: quietly, and without any guarantee. Give it two minutes. If you don't feel anything, you can come back and tell me I was wrong. I'll take the hit. But I don't think I'm wrong. --- 32/ A PREDICTION By the end of the week, this video will be in your group chat. Someone will send it with no caption, just three exclamation marks. Someone else will reply "WHAT." A third person will say "wait, is that real?" and someone who has already watched it will say "watch the whole thing." That's how the good ones travel. Not through algorithms. Through people saying to each other: you have to see this. So. I'm saying it. You have to see this. --- 33/ IF YOU'RE STILL READING You made it this far, which tells me one of two things. Either you're already curious, or you're the kind of person who reads long posts to the end. Both of those people will love this video. Here's what I'd suggest. Scroll up. Press play. Turn the sound on. Put the phone in both hands, like you're holding something that might wake up. And when it's over, come back here and tell me one thing. Not "wow." Not "amazing." Tell me the exact second you stopped breathing. Everybody has one. I'll tell you mine if you tell me yours. --- 34/ WHY I'LL WATCH IT AGAIN TOMORROW Most videos, I watch once and forget by dinner. A few, I watch and keep. This is one I keep. Not for the shock. The shock fades. I keep it because it reminds me that the world still has moments you can't script, can't fake and can't predict: moments where something wild and something gentle end up in the same place at the same time, and the outcome depends entirely on a few quiet decisions. I keep it because it makes me want to be the kind of person who stays calm when it counts. And I keep it because every time I rewatch it, I notice one small thing I missed. You will too. --- 35/ THE LAST THING Somewhere out there right now, there is a wild animal in a place it shouldn't be, running out of options, deciding whether the nearest human is a threat or a chance. Most of the time, nobody is there. Sometimes somebody is. And sometimes, once in a long while, that somebody does the quiet, careful, unglamorous thing that turns a frightening moment into a story worth telling. That's what you're about to watch. Not a bird in a car. A very large, very wild thing that made a very small, very human choice possible. Press play. Then send it to someone. Then tell me what you saw. P.S. If this post made you curious, you're exactly who the video was made for. If it didn't, watch it anyway. Ten seconds in, you'll understand why I wrote 25,000 characters about a moment that most people would have tried to explain in one. Repost it for the person who needs to see it today.
2:17

A wild eagle got into his car. That's not the strange part. The strange part is that it didn't want to leave. Read that again, because it breaks everything you think you know about wild animals. An eagle is not a pigeon. It is not a stray dog. It is not a cat that wandered into your garage. It is a predator built over millions of years to avoid exactly one thing: being close to you. And yet. I watched this video more times than I'd admit. Not because it's cute. Not because it's scary. Because somewhere around the second minute, I realized I was holding my breath, and couldn't tell you why. So let me tell you what this post is about, and what it is NOT about. It is not about a bird in a car. It is about what happens in the quiet space between a wild animal and a human being when the animal has run out of options and chooses to stop running. Let's go. --- 1/ THE RULE EVERYONE BREAKS There is one rule in wildlife work. People who spend their lives around raptors repeat it like a prayer: A healthy wild bird of prey does not come to you. Not for food. Not for warmth. Not for shelter. Not ever. An eagle that lets you walk up to it is not brave. It is not tame. It is not "friendly." It is a bird that has decided the danger in front of it is smaller than the danger it was already in. Think about what that means. Think about how bad things have to get for the apex predator of the sky to look at a human being and quietly decide: "Okay. You. Maybe you." Most people never see that moment. They hear about it later, in a sentence that starts with "apparently, this bird just..." This time somebody was rolling. --- 2/ WHY YOU CAN'T LOOK AWAY (AND IT'S NOT AN ACCIDENT) Your brain has a built-in alarm for two things at the same time: power and vulnerability. Show it a lion, it goes quiet and sharp. Show it a baby bird, it softens. Show it both in the same body and something short-circuits. That's what an injured eagle is. A creature that looks like a weapon and behaves like a plea. The talons are enormous. The beak is a hooked blade. The eyes have a stare most humans can't hold for more than a few seconds. Everything about the animal says: stay away. And everything about its behavior says: please don't. That contradiction is the entire video. You will feel it in your chest before you can name it. Most of us don't have a word for this feeling because we almost never get to have it. --- 3/ THE SIZE PROBLEM Quick exercise. Before you watch, try to hold a number in your head. How big do you think an eagle is? Most people say "like a big chicken." Maybe a large cat. Then you see one inside a normal car, on a normal seat, next to normal car things like seat belts and a steering wheel, and your brain refuses the scale. The car suddenly looks small. The man suddenly looks small. And there is a moment in this video where you will do a thing nearly everyone does: you will pause, lean back from your screen, and say "no way that's real." It's real. --- 4/ WHAT THE MAN DID NEXT (WILL SURPRISE YOU) I'm not going to tell you. I'll tell you what I expected, though. I expected panic. I expected screaming, a door flung open, a phone dropped, a very short video that ends with a blur and a swear word. That's the internet's favorite genre. You know it. A person encounters an animal, loses their mind, the clip is 8 seconds long, everyone laughs. This is not that clip. This clip runs a little over two minutes. And it holds you the whole way. Because what happens inside that car is less like a funny video and more like watching somebody defuse something. Slow. Careful. Almost polite. You'll catch yourself whispering at your screen. You'll catch yourself telling a stranger what to do. --- 5/ THE SOUND Turn the volume on. I'm serious. Not "recommended." Required. Half of what makes this video land is not what you see. It's what you hear: a man who is clearly, absolutely, completely out of his depth, talking his way through it anyway, one half-joke and one swear word at a time. Humor is what people do when the situation is bigger than they are. Somewhere in there is a lesson that has nothing to do with eagles. --- 6/ THE PART NOBODY TALKS ABOUT Everybody who comments on this video says the same three things: "Wow, the eagle is huge." "Wow, that guy is calm." "Wow, I would've screamed." Almost nobody says the fourth thing, which is the only one that matters: It's a rescue. Not a rescue of the bird. Well, yes, of the bird. But more precisely, it's a rescue that only works because two completely different kinds of creature agree, for a few minutes, to not do what their instincts are screaming at them to do. One of them is built to strike. The other is built to flee. Neither does. If you've ever wondered what trust looks like when neither side has words, this is the closest thing I've ever seen on a screen. --- 7/ THE THING ABOUT STILLNESS We live in a world that rewards motion. Post more. Reply faster. Move. Pivot. Hustle. Scale. And then there's a video where the most powerful thing in the entire frame is a creature sitting completely still, and the second most powerful is a man doing his best to match it. Nobody is performing. Nobody is trying to go viral. (Which is exactly why it will.) No sponsored segment. No "wait for it." Just stillness, and tension, and a decision being made in real time by two living things that can't explain themselves to each other. I've been on this platform for over a decade. I've seen what spreads. And it's almost never the loud stuff. It's the stuff that makes your whole feed go quiet for a second. This does that. --- 8/ SEVEN THINGS I LEARNED ABOUT EAGLES AFTER THIS VIDEO BROKE MY BRAIN I went down a rabbit hole. Here's what I came back with. → An eagle's eyesight is estimated to be several times sharper than a human's. It can pick out a rabbit from a distance at which you couldn't tell a person from a post. It sees in a way you simply do not. When it looks at you from close range, it is not "looking." It is reading. → Their grip is brutal. The talons are built to lock, and the locking mechanism works with the bird's own weight. In other words, the harder it settles, the harder it holds. A bird like that does not need to try to be dangerous. Dangerous is the default setting. → Raptors hide pain. This is the one that got me. In the wild, a bird that shows weakness gets challenged, robbed or eaten. So they do the opposite of what we do: the worse they feel, the quieter and stiller they become. A calm eagle is not always a healthy eagle. → Many birds of prey are surprisingly light for their size. Huge wingspan, hollow-boned frame, almost nothing underneath all those feathers. The "giant" you see is mostly an illusion built from wings. Which makes the next part even stranger. → They are not social with humans. At all. Raptors are solitary hunters. There is no instinct in there that says "find the nearest person and sit next to them." If an eagle is near you and not leaving, something has gone wrong, or something has gone very, very right. → Eagles live long lives, and they hold territory with an almost military seriousness. To end up far from where it belongs, inside a machine it doesn't understand, surrounded by smells it has never encountered, is as far outside its world as a bird can be pushed. → And the last one: they don't do "sorry" and they don't do "thank you." There is no grateful look. If you're waiting for the moment where the bird bows its head, you'll be waiting forever. That's why the ending of this video hits the way it does. I won't say more. --- 9/ THE FIRST TEN SECONDS If you only give this video ten seconds, you'll understand it. If you give it a full minute, you'll have a hard time leaving. Ten seconds is enough because the whole story is in the frame at once: the scale, the closeness, the stillness, the human who has just realized that his day has changed. You don't need context. You don't need a caption. You don't need me. And that is the highest compliment I know how to give a video on this platform. Most content needs an explanation. This content needs a witness. Be the witness. --- 10/ WHY THE COMMENTS ARE GOING TO BE WILD I can already tell you what's coming, because I've seen this exact pattern on a thousand posts: First wave: "FAKE." Second wave: "It's trained." Third wave: "Why didn't he just open the door?" Fourth wave: someone who actually works with birds, calmly explaining what's going on in a way that makes everyone else feel slightly embarrassed. Fifth wave: a quiet reply that just says "I'm not crying, you're crying." If you're going to join the conversation, at least watch to the end first. The end is the answer to about 80% of the arguments. --- 11/ THE QUESTION THAT WON'T LEAVE ME ALONE I keep thinking about one thing. Why a car? Out of all the places in a whole landscape, why that one? Maybe it was open. Maybe it was dark. Maybe it was warm. Maybe it was the only enclosed space for miles. Maybe the bird was hurt and needed to be somewhere with walls. Maybe it was a coincidence. Here's what I think. And I'm saying "think," not "know." I think when a creature is hurt badly enough, it stops choosing the best place and starts choosing the nearest one. Not the safest. Not the smartest. The nearest. And the nearest thing happened to be a man, a seat, a roof, and a few feet of air that felt, to a frightened animal, like the closest thing to a cave. If that's true, it's a very uncomfortable idea. Because it means the bird didn't pick him. It means he was just the one who was there. And the whole story, everything after, depends on what he decided to do with that. --- 12/ THE MOMENT YOU THINK YOU KNOW WHAT'S COMING There's a beat in the video where you will be absolutely certain you know where this is going. You'll feel your brain build the ending. It will feel obvious. It will be wrong. I won't say how. But I'll say this: watch the person, not the bird. The bird is the headline. The person is the story. The more I watched, the more I noticed that the camera was never really on the eagle. It was on the human who had to decide what kind of person he was going to be in the next five minutes. That's a good test for anything you watch online: Is this video about the thing in the frame? Or is it about the choice someone is making just outside it? The best ones are always the second kind. --- 13/ A NOTE FOR PEOPLE WHO THINK THIS IS "JUST A CUTE ANIMAL VIDEO" It isn't. Cute animal videos are interchangeable. Puppy falls over. Cat stares at a wall. Hamster eats a tiny burrito. You scroll past, you smile, you forget. This is the other end of the spectrum. This video is not cute. This video is tense. It's patient. It's got weight. It sits in your chest for a while after it ends. If you want to know whether a clip is actually good, ask yourself one question an hour after you watch it: Can I still see it? I can still see this one. --- 14/ THE THING ABOUT BEING THE ONLY WITNESS Imagine this happening and nobody filming. A man. A car. An eagle. A decision. Nobody would believe it. He'd tell the story at a dinner and people would nod politely and think: sure. The camera changes everything. Not because it makes the story true, but because it makes the story shareable. And shareable is how good things survive. So here's what I ask. If you watch this and it gets to you, don't just like it and move on. Send it to one person who needs a reason to believe that the world still does unexpected, gentle things. You know who that person is. --- 15/ THE CONTRAST THAT MAKES IT WORK Great stories run on contrast. Always. Predator and prey. Strength and fragility. Speed and stillness. Fear and calm. A wild creature and a very ordinary place. This video stacks all of it. An animal that belongs to the open sky, pressed into the smallest space it has probably ever occupied. A human being whose biggest worry that morning was probably something like traffic or a phone battery. A moment that should be a disaster and keeps refusing to become one. When those contrasts sit that close together, they stop being opposites. They start being a conversation. It's the same reason a candle looks brighter in a dark room. --- 16/ WHAT YOUR BRAIN IS DOING WHILE YOU WATCH For the nerds. I love the nerds. Three things are happening in your head while this plays. One: threat detection. You are looking at a large predator in a confined space, and a very old part of your brain is running a calculation about how many seconds it would take to get out. It's doing this whether you want it to or not. Two: empathy. A second system, the one that makes you wince when you see someone stub a toe, is reading the posture of an animal that is clearly not at its best. It's asking whether the creature is okay. Three: prediction. A third system is trying to guess what happens next, over and over, every second, and the video keeps making that guess slightly wrong. Three engines, running together, pulling in different directions. That's why you can't scroll. Your brain is busy. --- 17/ "BUT WHAT DID IT FEEL LIKE TO BE THERE?" I'll be honest. I don't know. I've never had a wild raptor sitting behind me while I tried to work out what to do with my hands. But I've thought about it a lot, and I think it feels like the opposite of a movie. In a movie, the music tells you what to feel. You get the swell, the cut, the close-up. In real life, there's nothing. Just air. And a thing that's alive in a way you can't negotiate with, in the same space as you, and a clock you can't see. Nobody tells you what to do. You just have to do something. That's the part the video gets right without even trying. You are not watching a performance. You are watching someone make it up as they go. --- 18/ THE ENDING (I'M NOT GOING TO SPOIL IT, BUT I'M GOING TO TELL YOU HOW IT FEELS) You know the feeling at the end of a long flight, when the wheels touch the ground and the whole cabin exhales? It's that. You don't know you've been tense until it stops. And then, a second later, something quieter shows up. Not happiness exactly. More like relief with a little bit of awe mixed in, the feeling you get when you realize you just watched something that was not supposed to go this well. I've been on the internet a long time. That feeling is rarer than it should be. If you've been scrolling through outrage for the last hour, give yourself two minutes of this. Your nervous system will thank you. --- 19/ WHY THIS IS THE ANTIDOTE Open any feed right now and count how many posts are trying to make you angry. Ten? Twenty? Now count how many are trying to make you feel something steady and good about the world, without pretending that the world is simple. This video is one of those. And the reason it works is that it doesn't ask you to ignore anything. It doesn't tell you everything is fine. It shows you a situation where things could have gone wrong in at least five different ways, and then it shows you what happens when the people involved choose patience. That is not a naive message. That is the hardest message there is. --- 20/ THE PEOPLE YOU DON'T SEE Here's a thing that bothers me about viral videos. We always celebrate the person who stands in front of the camera. We almost never celebrate the people behind them: the ones who show up when called, who do the slow and unglamorous work, who carry what needs carrying, who stay with something frightened until it isn't. Wildlife rescue is almost entirely made of those people. Volunteers. Rangers. Rehabilitators. People who pick up the phone at strange hours and say "okay, I'm on my way," and then drive somewhere with a carrier and a towel and no guarantee of a good outcome. If this video makes you feel something, let it make you feel something about them too. They're the reason a lot of stories like this have an ending at all. --- 21/ THE LINE I KEEP REPEATING "A wild animal that stops running is asking you a question." I wrote that down after my third viewing, and I've been turning it over ever since. Because most of us, when we meet something frightening, ask ourselves: how do I get away? Almost none of us ask: what is it trying to say? The man in this video asks the second question. Whether he knows it or not. That's the difference between a clip that makes you gasp and a clip that makes you think. --- 22/ A SMALL RULE FOR YOUR OWN LIFE If you ever find yourself near an injured wild animal, here's the unglamorous, boring, correct advice that every rehabilitator will give you: Don't be a hero. Keep your distance. Keep your voice low. Keep your hands to yourself. Don't pick it up, don't feed it, don't try to fix it on your own, and call the people whose entire job is to know what to do. Wild animals can be hurt by well-meaning help, and they can hurt you, too. I say this because I know exactly what a video like this makes people want to do. It makes people want to rush in. Don't. Call someone who knows. That's the real heroism, and it's less photogenic, and it saves more lives. --- 23/ THE SENTENCE THAT CAME UP IN MY HEAD AT THE END "Some things only happen if you stay calm for long enough." I don't know if that's true. I know it's what I thought. I think it's true of eagles. I think it's true of people. I think it's true of most of the hard things we're afraid of: the conversation, the diagnosis, the email, the leap. The version of you that panics is not stupid. It's just early. If you can hold still just a little longer than your instincts want you to, the situation will usually show you something it was hiding. That's it. That's the thesis of the entire post. You didn't need an eagle for it. But an eagle makes it a lot harder to forget. --- 24/ FIVE THINGS TO WATCH FOR (ZERO SPOILERS) Pay attention to these, and the video becomes a different video. → The distance. Not between the man and the bird. Between the bird and everything else it could have done. → The hands. Not what they do, but what they don't do for as long as they possibly can. → The words. Every line is human, unpolished, and a little scared. That's the point. → The silence between the words. It's longer than you expect and it tells you more than the dialogue. → The change of scenery. The story doesn't end where you think it will. Stay for the second half. The second half is the reason this post exists. If you do that, you won't just watch it. You'll remember it. --- 25/ THE SCREENSHOT TEST Here's a trick I use to decide whether a video is worth sharing. Pause it at a random moment. Any moment. If the frozen image still tells a story, you have something special. If it's just a blur or a talking head, you have content. Pause this one anywhere. Anywhere at all. It's a painting. A creature that belongs to the sky, a very ordinary human world, and a tension between them so thick you could rest your hand on it. It works as a still. It works as a clip. It works at half speed. It works with the sound off, although you shouldn't do that to yourself. That's rare. Videos built on real moments have it. Videos built on engineering never do. --- 26/ A PROMISE I'm not selling anything. No course. No link in bio. No thread of "10 lessons from an eagle for your startup." (Okay, one of those earlier sections came dangerously close. I'm aware.) I just watched something that made me stop, and in a feed full of things that make me scroll faster, that felt like a small miracle. So the promise is simple. If you give it your two minutes, you won't feel like you wasted them. If you do, say so. I can take it. --- 27/ WHAT A WILD ANIMAL TEACHES YOU ABOUT TIMING There's a rhythm to moments like this, and it isn't the rhythm of the internet. The internet is fast. Cut, cut, cut. Hook, payoff, next. This video moves at the speed of something that's afraid and something that's trying not to add to it. Slowly. Carefully. With long pauses where nothing seems to happen and everything is actually happening. That's why it's so hard to describe and so easy to feel. A fast video asks for your eyes. A slow video asks for your pulse. This one gets it. --- 28/ IF YOU ONLY REMEMBER ONE THING Don't remember the size of the bird. Don't remember the car. Remember that for a few minutes, a frightened wild animal and a startled human being shared a very small space, and neither one made it worse. In a year when it feels like everybody is making everything worse, that's worth two minutes of your life. Watch it. Then watch the people around you a little differently for the rest of the day. --- 29/ ONE MORE THING BEFORE YOU PRESS PLAY Put the phone down in a quiet room if you can. Not in a queue. Not in a lift. Not between two meetings. Give it the space the animal asked for. --- 30/ WHO THIS VIDEO IS FOR If you've ever felt like the world is getting louder and colder, this is for you. If you've ever been afraid of something and done the right thing anyway, this is for you. If you've ever helped something that couldn't thank you, this is for you. If you grew up looking at the sky and wondering what it would be like to be that free, and that strong, and that far from everything, this one is definitely for you. And if you're the kind of person who says "I don't have time for a two-minute video," I get it. I really do. But I'd bet money that you'll watch it twice. --- 31/ THE REAL REASON I'M POSTING THIS I could have written a 280-character caption. A joke. A "wait for it." A little emoji. I wrote this instead because some videos deserve a slower introduction. Not because they need explaining. Because they need you to arrive in the right mood. You can't watch this while half-distracted. You can't watch it while composing a reply. You can't watch it while thinking about what you'll post next. It asks for your attention the way the animal asks for space: quietly, and without any guarantee. Give it two minutes. If you don't feel anything, you can come back and tell me I was wrong. I'll take the hit. But I don't think I'm wrong. --- 32/ A PREDICTION By the end of the week, this video will be in your group chat. Someone will send it with no caption, just three exclamation marks. Someone else will reply "WHAT." A third person will say "wait, is that real?" and someone who has already watched it will say "watch the whole thing." That's how the good ones travel. Not through algorithms. Through people saying to each other: you have to see this. So. I'm saying it. You have to see this. --- 33/ IF YOU'RE STILL READING You made it this far, which tells me one of two things. Either you're already curious, or you're the kind of person who reads long posts to the end. Both of those people will love this video. Here's what I'd suggest. Scroll up. Press play. Turn the sound on. Put the phone in both hands, like you're holding something that might wake up. And when it's over, come back here and tell me one thing. Not "wow." Not "amazing." Tell me the exact second you stopped breathing. Everybody has one. I'll tell you mine if you tell me yours. --- 34/ WHY I'LL WATCH IT AGAIN TOMORROW Most videos, I watch once and forget by dinner. A few, I watch and keep. This is one I keep. Not for the shock. The shock fades. I keep it because it reminds me that the world still has moments you can't script, can't fake and can't predict: moments where something wild and something gentle end up in the same place at the same time, and the outcome depends entirely on a few quiet decisions. I keep it because it makes me want to be the kind of person who stays calm when it counts. And I keep it because every time I rewatch it, I notice one small thing I missed. You will too. --- 35/ THE LAST THING Somewhere out there right now, there is a wild animal in a place it shouldn't be, running out of options, deciding whether the nearest human is a threat or a chance. Most of the time, nobody is there. Sometimes somebody is. And sometimes, once in a long while, that somebody does the quiet, careful, unglamorous thing that turns a frightening moment into a story worth telling. That's what you're about to watch. Not a bird in a car. A very large, very wild thing that made a very small, very human choice possible. Press play. Then send it to someone. Then tell me what you saw. P.S. If this post made you curious, you're exactly who the video was made for. If it didn't, watch it anyway. Ten seconds in, you'll understand why I wrote 25,000 characters about a moment that most people would have tried to explain in one. Repost it for the person who needs to see it today.

Earth Unveiled

152,403 views • 28 days ago

New article: WTF Is a Loop? Part 2: The 15 Loops People Are Actually Running (and the Commands to Steal Them) - now with video and music! (SOUND ON!). Part 1 received 3.6M views. Enjoy the ACTUAL loops you can use.
1:40

New article: WTF Is a Loop? Part 2: The 15 Loops People Are Actually Running (and the Commands to Steal Them) - now with video and music! (SOUND ON!). Part 1 received 3.6M views. Enjoy the ACTUAL loops you can use.

Matt Van Horn

31,262 views • 4 months ago

Loops vs. Graphs, clearly explained! loops are great, but they have a ceiling: a loop makes one unit of work better. it cannot decide which units exist. so you end up with a very good agent running the wrong three steps, in the wrong order, one at a time. Graph engineering fixes this by moving the decision up a layer: what runs, what runs at the same time, and what never runs at all. you need both. here's how it works: a graph splits your system into two kinds of decision. ↳ inside a unit: the loop. produce, check, correct, repeat until green ↳ between units: the graph. split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs you get parallel work, isolated contexts, and steps that stop running when nothing needs them. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. a graph where every edge is an agent pays rent on its own wiring. one thing to know before you scale it. a graph has two return paths, and almost everyone builds one. ↳ the correction edge is short. a gate rejects one unit back to the step that produced it, and it fixes the run you are in ↳ the learning edge is long. an accepted result goes back to the splitter as a constraint, and it fixes every run after skip the second and you get a graph that is fast and never gets smarter. next week it starts from the same place with the same blind spots. and a smaller one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. 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 ↓
0:41

Loops vs. Graphs, clearly explained! loops are great, but they have a ceiling: a loop makes one unit of work better. it cannot decide which units exist. so you end up with a very good agent running the wrong three steps, in the wrong order, one at a time. Graph engineering fixes this by moving the decision up a layer: what runs, what runs at the same time, and what never runs at all. you need both. here's how it works: a graph splits your system into two kinds of decision. ↳ inside a unit: the loop. produce, check, correct, repeat until green ↳ between units: the graph. split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs you get parallel work, isolated contexts, and steps that stop running when nothing needs them. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. a graph where every edge is an agent pays rent on its own wiring. one thing to know before you scale it. a graph has two return paths, and almost everyone builds one. ↳ the correction edge is short. a gate rejects one unit back to the step that produced it, and it fixes the run you are in ↳ the learning edge is long. an accepted result goes back to the splitter as a constraint, and it fixes every run after skip the second and you get a graph that is fast and never gets smarter. next week it starts from the same place with the same blind spots. and a smaller one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. 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

73,867 views • 2 months ago

"Courage is far shorter in supply than genius." - Peter Thiel "One of the challenges in writing a book about entrepreneurship or teaching a class on this is that there is sort of no formula. And I think science always starts with a number two. It starts with experiments you can repeat, things you can do over and over again. But there's sort of a sense in which every moment in the history of business, every moment in the history of technology happens only once. The next Mark Zuckerberg will not be starting a social networking company. The next Larry Page will not start a search engine. The next Bill Gates will not be starting an operating system. And so if you are trying to copy these people, you're in some sense not learning from them. And so I think one of the really big challenges in teaching or writing about entrepreneurship is what can you say about being an entrepreneur at all when the key thing is always to do something new, different, that's not precisely been done before. And so the point of departure I start with in Zero to One is a somewhat indirect approach by asking a series of contrarian questions. The business question is, what great company is nobody starting? The more intellectual version of this question is, tell me something true that very few people agree with you on. And this is a fantastic interview question. It turns out to be quite a hard question, even when people can read on the internet that you ask of everybody who comes in the door, it still is a hard question. It's one of those unusual questions where if you know it's on the test, it's still hard. And it's hard not just because we sort of think that new things require brilliance or something like that, but because it's socially difficult. If I ask you that question, if you tell me something like the education system is screwed up or our political system doesn't work very well, those are true answers, but they're not actually good answers because all of us already know them to be true. The good answers are ones that are somehow uncomfortable that the person interviewing you does not actually want to hear. And I think we live in this world where courage is in far shorter supply than genius. And so it is sort of this, it is in a sense this problem of political correctness properly understood, is this very deep, very, very broad sort of a problem."
2:28

"Courage is far shorter in supply than genius." - Peter Thiel "One of the challenges in writing a book about entrepreneurship or teaching a class on this is that there is sort of no formula. And I think science always starts with a number two. It starts with experiments you can repeat, things you can do over and over again. But there's sort of a sense in which every moment in the history of business, every moment in the history of technology happens only once. The next Mark Zuckerberg will not be starting a social networking company. The next Larry Page will not start a search engine. The next Bill Gates will not be starting an operating system. And so if you are trying to copy these people, you're in some sense not learning from them. And so I think one of the really big challenges in teaching or writing about entrepreneurship is what can you say about being an entrepreneur at all when the key thing is always to do something new, different, that's not precisely been done before. And so the point of departure I start with in Zero to One is a somewhat indirect approach by asking a series of contrarian questions. The business question is, what great company is nobody starting? The more intellectual version of this question is, tell me something true that very few people agree with you on. And this is a fantastic interview question. It turns out to be quite a hard question, even when people can read on the internet that you ask of everybody who comes in the door, it still is a hard question. It's one of those unusual questions where if you know it's on the test, it's still hard. And it's hard not just because we sort of think that new things require brilliance or something like that, but because it's socially difficult. If I ask you that question, if you tell me something like the education system is screwed up or our political system doesn't work very well, those are true answers, but they're not actually good answers because all of us already know them to be true. The good answers are ones that are somehow uncomfortable that the person interviewing you does not actually want to hear. And I think we live in this world where courage is in far shorter supply than genius. And so it is sort of this, it is in a sense this problem of political correctness properly understood, is this very deep, very, very broad sort of a problem."

Founder Mode

12,339 views • 9 months ago