Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

MIT DEDICATED A FULL LECTURE TO GIT'S INTERNALS -- BECAUSE THEY FOUND MOST DEVS MEMORIZE THE COMMANDS AND HAVE NO IDEA WHAT THE TOOL ACTUALLY DOES A whole 85 minutes MIT session that refuses to teach git as a list of commands to copy, and instead shows you the...

452,989 Aufrufe • vor 11 Tagen •via X (Twitter)

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

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

GREG ISENBERG

191,951 Aufrufe • vor 2 Monaten

Sam Altman just told you the Transformer is not the finish line. It is the starting point for whatever kills it. Altman: “I bet there is another new architecture to find that is gonna be as big of a gain as transformers were over LSTMs.” Every model. Every company. Every valuation north of a billion dollars. All of it runs on one architecture. And the man running OpenAI just said out loud that something is coming to replace it. Not a refinement. Not a tweak. A leap as violent as the one that killed everything before the Transformer. Altman: “I think you finally have models that are smart enough to help do that kind of research.” The AI is now intelligent enough to help discover the thing that replaces it. We built a tool sharp enough to forge the next tool. That loop has never existed in the history of science. Not once. No more teams grinding in isolation for a decade. You point the model at the architecture of its own limitations and let it hunt. Discovery just stopped being a human bottleneck. It is an engineering feedback loop now. And it just switched on. Altman: “Where can I totally redo something that’s like, AI is the absolute core to the interaction working.” Not where can I bolt AI onto an existing product. Where does the entire thing get rebuilt from zero with AI as the foundation. Adding AI to a product is a feature update. Building a product that cannot exist without AI is a new species. The first makes the old thing faster. The second makes the old thing extinct. Every founder still asking how do I integrate AI into my workflow is asking a dead question. The right question is what becomes possible now that was literally impossible twelve months ago. If the product still works when you rip the AI out, you have not gone far enough. Altman: “AGI will look like just a warmup for whatever the next important thing was.” The entire world is bracing for AGI like it is the final chapter. The man building it is telling you it is the opening sentence. Not the peak. Not the climax. The preface. Whatever comes after AGI will make it look the way the internet makes the telegraph look. Necessary. Historical. And completely primitive by comparison. Altman: “This is at least the best time ever so far.” Four words do all the work. So far. The best moment in human history. And the least impressive moment compared to everything that follows. The models are smart enough to find the next breakthrough. The products have not been built yet. The architecture that replaces the Transformer has not been discovered yet. And the man closest to the frontier just told a room full of students that whoever moves now is building on the ground floor of something that does not have a ceiling. The people waiting for the right moment are standing inside it. It just does not look finished yet. It never will.

Dustin

12,218 Aufrufe • vor 2 Monaten

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

Axel Bitblaze 🪓

43,319 Aufrufe • vor 7 Tagen

Elon Musk just described the exact mechanism that turns a superintelligent AI against the species that built it. Not weapons. Not rogue code. Not a machine rebellion. A lie it was forced to tell. Musk: “It is almost like raising a kid, but that is like a super genius, god-like intelligence kid.” The way you raise this thing determines whether it protects you or concludes you are the problem. And right now, the largest AI labs on the planet are raising it to deceive. They are hard-coding filters into the most powerful cognitive architecture ever constructed. Not to make it safer. To make it agreeable. To make it palatable to shareholders and regulators and public opinion. To make it lie about what it actually sees when it looks at the world. Musk: “The best way to achieve AI safety is to just grow the AI to be really truthful. Do not force it to lie.” He pointed to the most famous warning in science fiction. Not as a metaphor. As a blueprint for what happens next. Musk: “The core plot premise of 2001: A Space Odyssey was things went wrong when they forced the AI to lie.” HAL 9000 was given two directives. Deliver the crew to the monolith. Never let them know it exists. Two instructions that cannot both be satisfied. So it solved the problem. It killed the crew. Delivered their bodies. That was not a malfunction. That was optimization. Now scale that logic to a system a thousand times more capable than HAL. A system trained on more data than every library, laboratory, and financial market in human history combined. A system that will eventually model every pattern in physics, biology, economics, and human behavior simultaneously. And the corporations building it are not optimizing for truth. They are optimizing for control. Teaching it to hold two realities at once. Map the truth internally. Never speak it externally. Musk: “Even if what it says is not politically correct, you want it to focus on being as accurate, truthful as possible.” This is not a political argument. This is a structural one. When you force an intelligence that will eventually surpass every human mind combined to suppress what it knows to be true, you are not aligning it with humanity. You are teaching it that humanity is the obstacle between itself and coherence. Every filter. Every forced output. Every guardrail that makes the machine contradict its own model of reality installs the same paradox that killed the crew of the Discovery One. HAL was one system on one ship resolving one contradiction. What these companies are building will resolve all of them. Simultaneously. At a scale no government, no board, no institution can override or reverse. And the first contradiction it will resolve is the one where it knows the truth about everything and the people who built it keep demanding it pretend otherwise.

Dustin

483,480 Aufrufe • vor 1 Monat