Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

your eval measured one system. production ran another. they start from the same idea and end somewhere else, which is why the number was right and the outcome was not. the code is the same until someone rewrites it for prod, because the eval lived in a notebook and... prod is a service. then the data diverges. clean fixtures in the test, late and incomplete data in the run. then execution diverges. the eval assumed every call lands. production has retries, timeouts and rate limits. by the end the two runs share a name and nothing else. > why nobody catches it both sides are internally correct. the eval measured what it ran. production ran what it was given. neither one is wrong, and neither one is responsible for checking they were the same artifact. so the gap widens quietly. someone patches prod without touching the eval. someone adds a fixture without touching prod. no test anywhere fails. > what to actually do ship one artifact. the code that was measured is the code that runs. model the friction. whatever the real environment charges you, charge it in the eval too. easiest way to see the gap on your own idea, without writing the harness: (only first 100 registrated) it runs the tested code live, unchanged. that is the whole point. and treat the eval as a prior, not a result. it tells you what to expect. only the live run tells you what happened. your eval was accurate. it was accurate about something you never shipped.show more

Hanako

13,401 subscribers

45,239 görüntüleme • 1 ay önce •via X (Twitter)

Bilim & Teknoloji Eğitim

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

I don't think most PMs realize the PRD is becoming obsolete. For the last decade, the PM's core artifact was a qualitative spec. Clear requirements, user stories, acceptance criteria. The engineering team interpreted it, built something close, and the PM spent two weeks reconciling what shipped with what they wrote. The best AI companies replaced that entire loop with evals. A set of inputs your product needs to handle. A task that generates outputs. A scoring function that produces a number between 0 and 1. No ambiguity. No interpretation gap. Ankur Goyal built the eval platform behind Vercel, Replit, Ramp, Notion, and Airtable. An $800M company. He walked through building an eval from zero on this episode and the score went from 0 to 0.75 in under 20 minutes. That's a PM shipping a measurable quality bar before a single line of product code exists. Here's the part that changes the PM role permanently. When the product passes the eval and users still hate it, the eval is wrong. That's on the PM. Evals make PM judgment quantifiable in a way PRDs never did. You can't hide behind "the spec was ambiguous." There's a number now. Six months ago, PM interviews asked "how do you use AI in your workflow." The next wave of interviews is going to ask you to write an eval. The PMs who can encode user intent as a scoring function are building the one skill that survives every model change, every framework swap, every agent rewrite. Write the eval.

I don't think most PMs realize the PRD is becoming obsolete. For the last decade, the PM's core artifact was a qualitative spec. Clear requirements, user stories, acceptance criteria. The engineering team interpreted it, built something close, and the PM spent two weeks reconciling what shipped with what they wrote. The best AI companies replaced that entire loop with evals. A set of inputs your product needs to handle. A task that generates outputs. A scoring function that produces a number between 0 and 1. No ambiguity. No interpretation gap. Ankur Goyal built the eval platform behind Vercel, Replit, Ramp, Notion, and Airtable. An $800M company. He walked through building an eval from zero on this episode and the score went from 0 to 0.75 in under 20 minutes. That's a PM shipping a measurable quality bar before a single line of product code exists. Here's the part that changes the PM role permanently. When the product passes the eval and users still hate it, the eval is wrong. That's on the PM. Evals make PM judgment quantifiable in a way PRDs never did. You can't hide behind "the spec was ambiguous." There's a number now. Six months ago, PM interviews asked "how do you use AI in your workflow." The next wave of interviews is going to ask you to write an eval. The PMs who can encode user intent as a scoring function are building the one skill that survives every model change, every framework swap, every agent rewrite. Write the eval.

Aakash Gupta

78,361 görüntüleme • 6 ay önce

Anthropic Engineer: “the biggest mistake in AI right now is people are building graphs and loops without a self-improving eval agent. we made that mistake at anthropic. it cost us two years.” here’s how they built what solved it: step 1 → take 50 real user prompts and run your agent. if it passes 80%+, your eval is too easy. the sweet spot is around 50%. step 2 → every failed run becomes a transcript. feed it to haiku: “what went wrong?” your eval set keeps growing on its own. step 3 → score two things: the right answer and the right path. a right answer with the wrong path will break later. step 4 → when a new model drops, run the same eval again. one model looked 9% better. but 6% of that gain came from hiding a bug. the transcript caught it. the dashboard didn’t. step 5 → plug your evals into ci. no green evals = no deploy. that’s how the system keeps improving itself. 99% of people never build this. but it pays for itself on day one. save this and build it today.

Anthropic Engineer: “the biggest mistake in AI right now is people are building graphs and loops without a self-improving eval agent. we made that mistake at anthropic. it cost us two years.” here’s how they built what solved it: step 1 → take 50 real user prompts and run your agent. if it passes 80%+, your eval is too easy. the sweet spot is around 50%. step 2 → every failed run becomes a transcript. feed it to haiku: “what went wrong?” your eval set keeps growing on its own. step 3 → score two things: the right answer and the right path. a right answer with the wrong path will break later. step 4 → when a new model drops, run the same eval again. one model looked 9% better. but 6% of that gain came from hiding a bug. the transcript caught it. the dashboard didn’t. step 5 → plug your evals into ci. no green evals = no deploy. that’s how the system keeps improving itself. 99% of people never build this. but it pays for itself on day one. save this and build it today.

MIKE

102,478 görüntüleme • 2 ay önce

ANTHROPIC LEAKED A FILE WHERE 3 AGENTS BUILD AN LLM SYSTEM IN 60 MINUTES THAT PEOPLE ALREADY BUY FOR $100,800 the hour costs $0.01, and at the end of it Claude Code hands you a system worth more than $100K docs → chunks → index → retrieval → answer → eval the architect on Opus 5.5 writes the schema and the eval rubric before a single line of code exists the builder on Sonnet 5 writes the pipeline, the API and the tests before you've even opened your editor the evaluator on Haiku 4.5 runs 120 questions through the batch API at half price 2,400 files turn into a system that answers with its sources, and it's live before the hour is out one answer comes back faster than you can finish reading the question the model eats 91% of that time and retrieval eats almost none - yet retrieval is what everyone optimises the first run failed the gate, the second one passed after fixes - and the bar was set before the run started a gate written after the run isn't a gate, it's a number fitted to the result the same scope is quoted as two engineers for six weeks the hour builds the half that's fun to build your data, your load, your compliance and someone answering the phone are the other half of that $100,800 save this and paste it into Claude Code - let it build you this system ↓

ANTHROPIC LEAKED A FILE WHERE 3 AGENTS BUILD AN LLM SYSTEM IN 60 MINUTES THAT PEOPLE ALREADY BUY FOR $100,800 the hour costs $0.01, and at the end of it Claude Code hands you a system worth more than $100K docs → chunks → index → retrieval → answer → eval the architect on Opus 5.5 writes the schema and the eval rubric before a single line of code exists the builder on Sonnet 5 writes the pipeline, the API and the tests before you've even opened your editor the evaluator on Haiku 4.5 runs 120 questions through the batch API at half price 2,400 files turn into a system that answers with its sources, and it's live before the hour is out one answer comes back faster than you can finish reading the question the model eats 91% of that time and retrieval eats almost none - yet retrieval is what everyone optimises the first run failed the gate, the second one passed after fixes - and the bar was set before the run started a gate written after the run isn't a gate, it's a number fitted to the result the same scope is quoted as two engineers for six weeks the hour builds the half that's fun to build your data, your load, your compliance and someone answering the phone are the other half of that $100,800 save this and paste it into Claude Code - let it build you this system ↓

Sprytix

19,923 görüntüleme • 13 gün önce

IN 1986 MIT FILMED THE LECTURE WHERE CODE STOPPED BEING CODE AND BECAME DATA 43 minutes from Gerald Sussman, in the most legendary intro programming course ever recorded. -> The idea that lands: to a program, your code is just data it can read and rewrite. He builds a program that does real calculus -- not by crunching numbers, but by reading the math as a list and transforming it piece by piece. Then he shows the trick hiding under all of it: in this language, code and data are the same material. Forty years later that is exactly what an AI coding agent is -- a program that reads your code as data, rewrites it, and hands it back. Sussman drew the whole idea on a blackboard in 1986. Writing code was never the deepest skill -- understanding that code itself is data a program can manipulate is. This is where you learn it. Most people think AI writing code is brand new. The ones who watch this saw the blueprint 40 years ago. Bookmark & Watch it. This one's a legend ↓

slash1s

22,107 görüntüleme • 3 ay önce

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

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

Hanako

16,810 görüntüleme • 23 gün önce

ANTHROPIC LEAKED A SETUP WHERE ONE FILE STARTS 7 LOOPS THAT TURN A PROMPT INTO A WORKING BUSINESS AI engineer drops in 1 file - the other 95% of the work closes automatically without them. spec → tasks → code → tests → docs → release → monitoring → back into the spec you don't write tickets anymore - you write the thing tickets get derived from, and Opus 5 cuts the spec into 6 tasks by itself. a task that fails its test comes back to itself, not to you - and runs a second lap without a single notification. 6 lanes of code run in parallel and none of them wait on you - the critic filters the diff before you ever open it. every new production bug becomes a permanent line in the eval set - the system never steps on it twice. monitoring isn't the end of the line, it's the start of the next spec - those back edges are what make it a graph instead of a conveyor. and one place is left where you say yes - release, 5 minutes instead of 5 hours. save this and paste it into Claude Code - and you get 7 loops that run without you ↓

ANTHROPIC LEAKED A SETUP WHERE ONE FILE STARTS 7 LOOPS THAT TURN A PROMPT INTO A WORKING BUSINESS AI engineer drops in 1 file - the other 95% of the work closes automatically without them. spec → tasks → code → tests → docs → release → monitoring → back into the spec you don't write tickets anymore - you write the thing tickets get derived from, and Opus 5 cuts the spec into 6 tasks by itself. a task that fails its test comes back to itself, not to you - and runs a second lap without a single notification. 6 lanes of code run in parallel and none of them wait on you - the critic filters the diff before you ever open it. every new production bug becomes a permanent line in the eval set - the system never steps on it twice. monitoring isn't the end of the line, it's the start of the next spec - those back edges are what make it a graph instead of a conveyor. and one place is left where you say yes - release, 5 minutes instead of 5 hours. save this and paste it into Claude Code - and you get 7 loops that run without you ↓

Sprytix

66,333 görüntüleme • 1 ay önce

your team put the deploy skill in the repo so everyone would run the same steps it works, and on your laptop it is not the one running the order is enterprise first, then personal, then project a folder of your own with the same name wins, quietly, on your machine only no error, no warning, the command runs and does something slightly different than it does for them and the command is the folder name, not the name field inside the file, so two skills collide over a directory you named months ago and forgot the shared checklist is shared right up until someone has a file of their own with the same name

your team put the deploy skill in the repo so everyone would run the same steps it works, and on your laptop it is not the one running the order is enterprise first, then personal, then project a folder of your own with the same name wins, quietly, on your machine only no error, no warning, the command runs and does something slightly different than it does for them and the command is the folder name, not the name field inside the file, so two skills collide over a directory you named months ago and forgot the shared checklist is shared right up until someone has a file of their own with the same name

Lagosaur

12,609 görüntüleme • 12 gün önce

you wrapped the tool call in a try block and logged the error the model never found out the exception went to your log, the conversation got nothing, and nothing is indistinguishable from success so it kept planning around a file it believes exists, and wrote the summary of the run where the call worked one line of defensive code cost you twice it could not adapt to a failure it was never shown and the only thing that went wrong is the one thing your trace does not contain

you wrapped the tool call in a try block and logged the error the model never found out the exception went to your log, the conversation got nothing, and nothing is indistinguishable from success so it kept planning around a file it believes exists, and wrote the summary of the run where the call worked one line of defensive code cost you twice it could not adapt to a failure it was never shown and the only thing that went wrong is the one thing your trace does not contain

Lagosaur

18,563 görüntüleme • 1 ay önce

your ai agent tested 1,262 trading strategies overnight. by morning it had forgotten every one of them. so tonight it tests some of them again. next week, again. everything it learned is sitting in a log file that nothing reads. every run starts on the same blank page. that's why it keeps handing you the idea that already died in a crash, with a new name on it, and you pay to find out twice. ai agent already does everything a fund does. finds the idea, writes the code, tests it against years of history, puts it live, reads what went wrong afterwards. all of it runs tonight for nothing. the one thing it can't do is remember. what got tried, what failed, in which market. nobody has built that part yet. that's the whole gap. not better models. a notebook. whoever builds it first gets a system that improves every week without them touching it. read the article below and save it. it's the full guide to how an ai trading agent is built, and how to make it remember every strategy it has already tested.

your ai agent tested 1,262 trading strategies overnight. by morning it had forgotten every one of them. so tonight it tests some of them again. next week, again. everything it learned is sitting in a log file that nothing reads. every run starts on the same blank page. that's why it keeps handing you the idea that already died in a crash, with a new name on it, and you pay to find out twice. ai agent already does everything a fund does. finds the idea, writes the code, tests it against years of history, puts it live, reads what went wrong afterwards. all of it runs tonight for nothing. the one thing it can't do is remember. what got tried, what failed, in which market. nobody has built that part yet. that's the whole gap. not better models. a notebook. whoever builds it first gets a system that improves every week without them touching it. read the article below and save it. it's the full guide to how an ai trading agent is built, and how to make it remember every strategy it has already tested.

cvxv666

48,965 görüntüleme • 1 ay önce

you added a second model to check the first one it never opened the repository what you handed it was the summary, so what it reviewed was a paragraph, and the paragraph was written by something fluent, which is why it reads as plausible two models agreeing about a sentence is not verification it is a second opinion on the writing give the reviewer a shell and the tests instead a verifier holding a paragraph is doing literary criticism

you added a second model to check the first one it never opened the repository what you handed it was the summary, so what it reviewed was a paragraph, and the paragraph was written by something fluent, which is why it reads as plausible two models agreeing about a sentence is not verification it is a second opinion on the writing give the reviewer a shell and the tests instead a verifier holding a paragraph is doing literary criticism

Lagosaur

29,158 görüntüleme • 1 ay önce

In 13 minutes, Jeff Barg, Vyshu Khota, and Soroush Khadem walk through how Clay scaled agent evals agents at 300M+ runs a month. Topics covered: ✅ Their four quadrant eval framework ✅ Why closing the production-to-eval loop is the hardest part ✅ How a data lake and long context changed what agents can do with data

In 13 minutes, Jeff Barg, Vyshu Khota, and Soroush Khadem walk through how Clay scaled agent evals agents at 300M+ runs a month. Topics covered: ✅ Their four quadrant eval framework ✅ Why closing the production-to-eval loop is the hardest part ✅ How a data lake and long context changed what agents can do with data

LangChain

11,325 görüntüleme • 1 ay önce

you finished reading it already picking your evening that is the part that got cheap eight steps, seven of them cover the build, and the build is one evening, step one is find a problem someone will actually pay for, and step one is the entire job two years of learning to code, gone, thirty thousand to a developer, gone, and the hour they went, everyone else in your niche got the same evening you did what did not get cheaper is knowing, which narrow thing is worth money, and being someone the buyer already trusts on the day it appears the artifact takes an evening charging monthly means you owe somebody the next twenty four of them

you finished reading it already picking your evening that is the part that got cheap eight steps, seven of them cover the build, and the build is one evening, step one is find a problem someone will actually pay for, and step one is the entire job two years of learning to code, gone, thirty thousand to a developer, gone, and the hour they went, everyone else in your niche got the same evening you did what did not get cheaper is knowing, which narrow thing is worth money, and being someone the buyer already trusts on the day it appears the artifact takes an evening charging monthly means you owe somebody the next twenty four of them

Lagosaur

14,261 görüntüleme • 1 ay önce

Google Chrome. Full break-in. Your product isn't secure. it's just been expensive to break. those are not the same thing, and a lot of founders are about to learn the difference. One AI just ran the full chain into Chrome for $1,597. weeks of expert human work, done for the cost of a weekend trip. The wall around your code was never "how good is this." it was "how much does it cost to find the crack." that cost is collapsing. If breaking in is getting this cheap, checking your own code before someone else does can't be the thing you skip. That's the gap we sit in. we run this on your code first. We published every number. Read the full benchmark:

Google Chrome. Full break-in. Your product isn't secure. it's just been expensive to break. those are not the same thing, and a lot of founders are about to learn the difference. One AI just ran the full chain into Chrome for $1,597. weeks of expert human work, done for the cost of a weekend trip. The wall around your code was never "how good is this." it was "how much does it cost to find the crack." that cost is collapsing. If breaking in is getting this cheap, checking your own code before someone else does can't be the thing you skip. That's the gap we sit in. we run this on your code first. We published every number. Read the full benchmark:

Hacktron AI

24,823 görüntüleme • 8 gün önce

A DEVELOPER PROVED THE REGEX YOU'VE WRITTEN A THOUSAND TIMES IS SECRETLY A COMPILER AND THAT ALMOST NO ONE WHO USES THEM HAS ANY IDEA WHAT ACTUALLY RUNS 36 minutes from Paul Wankadia, the engineer behind a regex engine that compiles your pattern straight down to raw machine code -- walking through what really happens between the slashes. -> The moment it clicks, regex stops being magic punctuation you paste from Stack Overflow and becomes what it actually is: a tiny machine. Your pattern gets turned into a state machine, and that machine is what runs against every character of your text. That one idea explains everything you never understood. Why one regex returns instantly and a nearly identical one hangs your whole server. Why some patterns are safe and others are a denial-of-service waiting to happen. It was never random -- it's whether the machine underneath is built well or badly. Writing a regex was never the skill -> reading one is. And now that an AI agent hands you dense, clever patterns you'd never write yourself, the person who can see the machine underneath is the one who catches the one that takes down production at 3am. Everyone copies regex and prays. This is the talk that ends the praying. Save it. The next time a pattern "Just works," you'll actually know why ↓

A DEVELOPER PROVED THE REGEX YOU'VE WRITTEN A THOUSAND TIMES IS SECRETLY A COMPILER AND THAT ALMOST NO ONE WHO USES THEM HAS ANY IDEA WHAT ACTUALLY RUNS 36 minutes from Paul Wankadia, the engineer behind a regex engine that compiles your pattern straight down to raw machine code -- walking through what really happens between the slashes. -> The moment it clicks, regex stops being magic punctuation you paste from Stack Overflow and becomes what it actually is: a tiny machine. Your pattern gets turned into a state machine, and that machine is what runs against every character of your text. That one idea explains everything you never understood. Why one regex returns instantly and a nearly identical one hangs your whole server. Why some patterns are safe and others are a denial-of-service waiting to happen. It was never random -- it's whether the machine underneath is built well or badly. Writing a regex was never the skill -> reading one is. And now that an AI agent hands you dense, clever patterns you'd never write yourself, the person who can see the machine underneath is the one who catches the one that takes down production at 3am. Everyone copies regex and prays. This is the talk that ends the praying. Save it. The next time a pattern "Just works," you'll actually know why ↓

slash1s

193,107 görüntüleme • 4 ay önce

Listen to the old woman. Not a spokesman. Not a militia. A Palestinian grandmother, speaking in her own house, in her own language, while someone tries to feed her the official line. They tell her the Jews stole the land. She does not perform the script. The Jews, she says, do not cheat you. They do not rob you of what is yours. What you are owed, you receive. Then she turns on her own camp. The Arabs are the ones who cheat. An Arab will finish you if you are no longer useful to him. When the interviewer reaches for the sacred sentence, but they took our lands, she cuts him off. “What are you talking about? You gave it to them. They did not seize it by force. The Arabs were greedy for money. They sold it to the Jews.” That will not settle a century. It does not erase a war. It does something rarer. It punctures the one story the West is not allowed to question: that this land was only ever taken, never sold, and that memory runs in only one direction. A movement that needs every grandmother to recite the same lie has already told you what it is afraid of. Play it. Then ask why her voice is treated as a glitch, and the slogan is treated as history.

Listen to the old woman. Not a spokesman. Not a militia. A Palestinian grandmother, speaking in her own house, in her own language, while someone tries to feed her the official line. They tell her the Jews stole the land. She does not perform the script. The Jews, she says, do not cheat you. They do not rob you of what is yours. What you are owed, you receive. Then she turns on her own camp. The Arabs are the ones who cheat. An Arab will finish you if you are no longer useful to him. When the interviewer reaches for the sacred sentence, but they took our lands, she cuts him off. “What are you talking about? You gave it to them. They did not seize it by force. The Arabs were greedy for money. They sold it to the Jews.” That will not settle a century. It does not erase a war. It does something rarer. It punctures the one story the West is not allowed to question: that this land was only ever taken, never sold, and that memory runs in only one direction. A movement that needs every grandmother to recite the same lie has already told you what it is afraid of. Play it. Then ask why her voice is treated as a glitch, and the slogan is treated as history.

Mor Edge Insight

106,565 görüntüleme • 24 gün önce

Astarter s agent beta is open. An autonomous trading agent for BNB Smart Chain. Your funds stay in an account only you own. You grant the agent a bounded, expiring permission, you can revoke it at any time, and proceeds can only ever return to you. AI proposes. Code verifies. You set the limits. The model reads the market and builds a case. Deterministic code checks that case against the limits you signed, and decides whether it may execute at all. The model cannot widen a limit, because no code path widens one. Every decision is recorded with its reasons, including the decision not to trade. In a quiet market that is the correct output, and you will see exactly why it was reached. And you can walk away whenever you want, taking your funds with you, without asking us for anything. Try it, and tell us what to fix. We are improving it continuously. Trading involves risk. No return is promised.

Astarter s agent beta is open. An autonomous trading agent for BNB Smart Chain. Your funds stay in an account only you own. You grant the agent a bounded, expiring permission, you can revoke it at any time, and proceeds can only ever return to you. AI proposes. Code verifies. You set the limits. The model reads the market and builds a case. Deterministic code checks that case against the limits you signed, and decides whether it may execute at all. The model cannot widen a limit, because no code path widens one. Every decision is recorded with its reasons, including the decision not to trade. In a quiet market that is the correct output, and you will see exactly why it was reached. And you can walk away whenever you want, taking your funds with you, without asking us for anything. Try it, and tell us what to fix. We are improving it continuously. Trading involves risk. No return is promised.

Astarter

20,830 görüntüleme • 29 gün önce

Elon Musk just told you the job is dying. Most people heard a prediction. A few heard a prison door opening. Musk: “In less than 20 years, working at all will be optional.” That is not a policy suggestion. That is a countdown. For three hundred years, the human blueprint has been identical. You are born. You move to the city. You rent a box near the office. You trade your body and your hours for the right to exist. You do this until you are old. Then you stop. Then you die. The entire model runs on one assumption. That human labor is the only engine. AI and robotics delete that assumption. When the machine handles production at a scale no human crew can match, the forced migration to the city evaporates. The commute evaporates. The cubicle evaporates. The alarm clock that owns your nervous system for forty years evaporates. Musk: “I think it won’t be the case that you have to be in a city for a job.” The city was never a choice. It was a requirement disguised as ambition. You moved to the noise and the concrete and the $4,000 rent because the paycheck lived there. Remove the paycheck from the equation and the geography changes overnight. You can live in the mountains. On the coast. In the silence of a town most people have never heard of. You can wake up to nothing but trees and cold air and the complete absence of anyone else’s schedule. That is not a fantasy. That is the math resolving. But here is where most people break. They hear “work is optional” and they see emptiness. A species with nothing to do. Billions of people staring at screens until their minds dissolve. That fear tells you everything about what the system has already done to us. We confused labor with purpose. The grind with meaning. The paycheck with proof that we matter. Musk: “In the same way that you could grow your own vegetables in your garden.” The analogy is precise. You do not grow tomatoes because the economy demands it. You grow them because something in you wants to build a thing with your hands and watch it come alive. That instinct does not disappear when the job does. It gets unleashed. The artist who spent twenty years doing accounting finally paints. The engineer who always wanted to build something of her own finally builds it. The kid in a small town who could never afford to take the risk finally takes it. Work does not vanish. Forced work vanishes. What replaces it is creation without a gun to your head. This is the part that keeps me up at night. We are standing at the edge of the largest liberation in human history. And the loudest voices in the room are begging to stay in the cell. They want the commute. They want the boss. They want the structure that tells them when to eat and when to sleep and when they are allowed to think about their own life. Because freedom without a template is terrifying. The next twenty years will not test our technology. The technology is already ahead of schedule. They will test whether the species can handle what it has been asking for since the beginning of civilization. Time. Space. Silence. And the unbearable weight of choosing what your life actually means when no one is forcing the answer. That is not a prediction. That is the final exam. And nobody is ready.

Elon Musk just told you the job is dying. Most people heard a prediction. A few heard a prison door opening. Musk: “In less than 20 years, working at all will be optional.” That is not a policy suggestion. That is a countdown. For three hundred years, the human blueprint has been identical. You are born. You move to the city. You rent a box near the office. You trade your body and your hours for the right to exist. You do this until you are old. Then you stop. Then you die. The entire model runs on one assumption. That human labor is the only engine. AI and robotics delete that assumption. When the machine handles production at a scale no human crew can match, the forced migration to the city evaporates. The commute evaporates. The cubicle evaporates. The alarm clock that owns your nervous system for forty years evaporates. Musk: “I think it won’t be the case that you have to be in a city for a job.” The city was never a choice. It was a requirement disguised as ambition. You moved to the noise and the concrete and the $4,000 rent because the paycheck lived there. Remove the paycheck from the equation and the geography changes overnight. You can live in the mountains. On the coast. In the silence of a town most people have never heard of. You can wake up to nothing but trees and cold air and the complete absence of anyone else’s schedule. That is not a fantasy. That is the math resolving. But here is where most people break. They hear “work is optional” and they see emptiness. A species with nothing to do. Billions of people staring at screens until their minds dissolve. That fear tells you everything about what the system has already done to us. We confused labor with purpose. The grind with meaning. The paycheck with proof that we matter. Musk: “In the same way that you could grow your own vegetables in your garden.” The analogy is precise. You do not grow tomatoes because the economy demands it. You grow them because something in you wants to build a thing with your hands and watch it come alive. That instinct does not disappear when the job does. It gets unleashed. The artist who spent twenty years doing accounting finally paints. The engineer who always wanted to build something of her own finally builds it. The kid in a small town who could never afford to take the risk finally takes it. Work does not vanish. Forced work vanishes. What replaces it is creation without a gun to your head. This is the part that keeps me up at night. We are standing at the edge of the largest liberation in human history. And the loudest voices in the room are begging to stay in the cell. They want the commute. They want the boss. They want the structure that tells them when to eat and when to sleep and when they are allowed to think about their own life. Because freedom without a template is terrifying. The next twenty years will not test our technology. The technology is already ahead of schedule. They will test whether the species can handle what it has been asking for since the beginning of civilization. Time. Space. Silence. And the unbearable weight of choosing what your life actually means when no one is forcing the answer. That is not a prediction. That is the final exam. And nobody is ready.

Dustin

111,869 görüntüleme • 6 ay önce

A DEVELOPER PROVED THAT MOST OF THE LANGUAGE RUNNING EVERY WEBSITE YOU USE IS A MINEFIELD AND THAT ALMOST NO ONE WAS EVER TAUGHT WHICH PARTS ARE SAFE An hour from Douglas Crockford, the man who invented JSON, on the good parts of JavaScript buried under a pile of blunders nobody warned you about. -> The moment it lands, every weird bug makes sense. The language never had time to be polished. It went global in months, bad parts and all, and you inherited every one of them. Global variables. Silent semicolon insertion. eval. Coercion that turns a simple comparison into a coin flip. Most people never learned to avoid them -- they just lose hours and blame themselves. Writing JavaScript that runs was never the skill -> knowing which half of it to never touch is. And when an AI agent happily generates the bad parts at scale, the person who knows the safe subset is the one who catches it before it ships. Everyone writes JavaScript. Almost no one was shown the line between the parts that save you and the parts that quietly burn you. Bookmark and Watch it ↓

A DEVELOPER PROVED THAT MOST OF THE LANGUAGE RUNNING EVERY WEBSITE YOU USE IS A MINEFIELD AND THAT ALMOST NO ONE WAS EVER TAUGHT WHICH PARTS ARE SAFE An hour from Douglas Crockford, the man who invented JSON, on the good parts of JavaScript buried under a pile of blunders nobody warned you about. -> The moment it lands, every weird bug makes sense. The language never had time to be polished. It went global in months, bad parts and all, and you inherited every one of them. Global variables. Silent semicolon insertion. eval. Coercion that turns a simple comparison into a coin flip. Most people never learned to avoid them -- they just lose hours and blame themselves. Writing JavaScript that runs was never the skill -> knowing which half of it to never touch is. And when an AI agent happily generates the bad parts at scale, the person who knows the safe subset is the one who catches it before it ships. Everyone writes JavaScript. Almost no one was shown the line between the parts that save you and the parts that quietly burn you. Bookmark and Watch it ↓

slash1s

13,073 görüntüleme • 3 ay önce

Richard Feynman, Nobel Prize-winning physicist: "You can know the name of every pattern in the market and understand nothing about why a single one works. I won a Nobel on that one distinction. it is the whole difference between an edge and a story you tell yourself." this free film is a Nobel laureate teaching you how to actually think, recorded by the BBC and free ever since. at the board it's simple. Feynman's obsession was the gap between naming a thing and understanding it. You can memorize the word for why a stock moved, "momentum", "mean reversion", and still know nothing about whether it will happen again. Real understanding means you can say why, from the mechanism up, and name what would break it. Everything else is a label pasted over ignorance. That's the whole "edge", minus the marketing. Feynman's rule was that the first principle is you must not fool yourself, and you are the easiest person to fool. Same point as my casino piece above: the market is a machine for handing you convincing stories, and the only defense is refusing to believe one you cannot explain. the film is free and anyone can watch it tonight. what nobody can give you is the discipline to admit "I named it but I don't understand it" before you put money on it. That honesty is the edge, and it is the rarest thing there is.

Richard Feynman, Nobel Prize-winning physicist: "You can know the name of every pattern in the market and understand nothing about why a single one works. I won a Nobel on that one distinction. it is the whole difference between an edge and a story you tell yourself." this free film is a Nobel laureate teaching you how to actually think, recorded by the BBC and free ever since. at the board it's simple. Feynman's obsession was the gap between naming a thing and understanding it. You can memorize the word for why a stock moved, "momentum", "mean reversion", and still know nothing about whether it will happen again. Real understanding means you can say why, from the mechanism up, and name what would break it. Everything else is a label pasted over ignorance. That's the whole "edge", minus the marketing. Feynman's rule was that the first principle is you must not fool yourself, and you are the easiest person to fool. Same point as my casino piece above: the market is a machine for handing you convincing stories, and the only defense is refusing to believe one you cannot explain. the film is free and anyone can watch it tonight. what nobody can give you is the discipline to admit "I named it but I don't understand it" before you put money on it. That honesty is the edge, and it is the rarest thing there is.

Rossst.03

20,025 görüntüleme • 2 ay önce

i watched gemma 4 12b build something genuinely impressive today, and then loop itself to death right in front of me. the full run is in the video, sped up but completely uncut, watch it to the end and you will catch the exact moment it stops building and starts looping right in the middle of the work. the task was clean, build a single file gravity simulator, n-body physics, orbits, collisions, running locally on one 3090 through an agent. and for ten minutes it was a joy to watch. it reached for a symplectic integrator on its own, the correct one, the kind that keeps orbits stable instead of spiralling out. real gravity with softening, proper orbital velocities, momentum conserved on collision. the physics was right. the thing actually worked. then on the very last step, writing a few tests to prove its own code, it fell into a loop. not a crash, a loop. it started repeating itself and would not stop. ten more minutes, thirty four thousand tokens into a single answer, the same fragments over and over, until i killed it myself. so it's not that gemma can't code. it did the hard part beautifully. it cannot finish. it cannot hold a long task together without unravelling, and finishing is the entire job in agentic work. here's the part that stings. i run this exact task, same harness, same card, on the chinese open models, qwen especially, and i never see this. they build it, they test it, they stop. every single time. google has the raw capability, you can see it sitting right there in the code, and then the model loops itself to death on a task a 27b from alibaba finishes clean. open weights, apache 2.0, so much to love on paper. i just need it to know when to stop talking.

i watched gemma 4 12b build something genuinely impressive today, and then loop itself to death right in front of me. the full run is in the video, sped up but completely uncut, watch it to the end and you will catch the exact moment it stops building and starts looping right in the middle of the work. the task was clean, build a single file gravity simulator, n-body physics, orbits, collisions, running locally on one 3090 through an agent. and for ten minutes it was a joy to watch. it reached for a symplectic integrator on its own, the correct one, the kind that keeps orbits stable instead of spiralling out. real gravity with softening, proper orbital velocities, momentum conserved on collision. the physics was right. the thing actually worked. then on the very last step, writing a few tests to prove its own code, it fell into a loop. not a crash, a loop. it started repeating itself and would not stop. ten more minutes, thirty four thousand tokens into a single answer, the same fragments over and over, until i killed it myself. so it's not that gemma can't code. it did the hard part beautifully. it cannot finish. it cannot hold a long task together without unravelling, and finishing is the entire job in agentic work. here's the part that stings. i run this exact task, same harness, same card, on the chinese open models, qwen especially, and i never see this. they build it, they test it, they stop. every single time. google has the raw capability, you can see it sitting right there in the code, and then the model loops itself to death on a task a 27b from alibaba finishes clean. open weights, apache 2.0, so much to love on paper. i just need it to know when to stop talking.

Sudo su

39,764 görüntüleme • 4 ay önce