Загрузка видео...

Не удалось загрузить видео

На главную

STANFORD BUILT AN LLM MEMORY THAT WORKS LIKE A BRAIN AND IT MADE STANDARD RAG LOOK ANCIENT no fine-tuning. no bigger model. just a smarter way to remember the insight: the brain doesn't store memories as isolated notes. it indexes them and links them - so one cue pulls...

N01ennn's profile picture

NO1ennn

4,065 subscribers

32,454 просмотров • 18 дней назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

MICROSOFT QUIETLY BUILT ONE MEMORY MODULE THAT PLUGS INTO ANY AI AGENT AND CUTS ITS CONTEXT BY UP TO 100X no retraining. no task-specific redesign. you just plug it in Microsoft's insight: agents don't need to remember what happened. they need the knowledge pulled from it - the facts and the reusable skills, not the raw logs so PlugMem stops storing history. it compiles experience into a knowledge graph, then feeds the agent only what matters the result: > up to 1-2 orders of magnitude fewer tokens > highest decision-utility per token of any system tested > one unchanged module beats task-specific memory on chat QA, multi-hop search, and web agents more memory was never the answer. better memory is - and Microsoft just shipped it as a plug-in bookmark this. the agents that win won't remember more, they'll remember smarter.
0:42

MICROSOFT QUIETLY BUILT ONE MEMORY MODULE THAT PLUGS INTO ANY AI AGENT AND CUTS ITS CONTEXT BY UP TO 100X no retraining. no task-specific redesign. you just plug it in Microsoft's insight: agents don't need to remember what happened. they need the knowledge pulled from it - the facts and the reusable skills, not the raw logs so PlugMem stops storing history. it compiles experience into a knowledge graph, then feeds the agent only what matters the result: > up to 1-2 orders of magnitude fewer tokens > highest decision-utility per token of any system tested > one unchanged module beats task-specific memory on chat QA, multi-hop search, and web agents more memory was never the answer. better memory is - and Microsoft just shipped it as a plug-in bookmark this. the agents that win won't remember more, they'll remember smarter.

NO1ennn

225,884 просмотров • 20 дней назад

THIS GUY CONNECTED HIS AI AGENTS TO HIS OBSIDIAN AND BUILT A BRAIN THAT LEARNS ON ITS OWN. HERE'S HOW TO BUILD IT Obsidian is just markdown files sitting in a folder. That turns out to be the perfect memory for an AI agent, because an agent can read and write those files directly. He wired his agents into the vault so they pull context from it, do the work, and write what they learned back. The notes aren't the point. The loop is, and it gets sharper every cycle How to build it: 1. Point an agent at your vault. The fastest way, no plugins, no API keys: open a terminal and run npx obsidian-mcp /path/to/your/vault. That exposes your Obsidian folder to Claude as a tool it can read, search, and write to. Add it to your Claude Code or Cowork config and restart 2. Confirm it can see the brain. Ask it: "list the notes in my vault and summarize what's in them." If it reads them back, the connection is live. Now it starts every task with everything the vault already holds instead of from zero 3. Give each agent one job and a write-back rule. Tell it: "research this, then save what you found as a new note in /brain with links to related notes." One agent researches, one summarizes, one plans. Each writes its output back into the vault 4. Close the loop. Add one line to every agent's instructions: "read /brain before starting, write your result back when done." Now each task leaves the vault richer, and the next run reads that before it works. It compounds instead of resetting 5. You only steer. Review what the brain produces, point it at the next thing. The agents handle the reading, writing, and connecting The edge isn't better notes. It's a brain that feeds itself, so the work gets sharper every cycle instead of starting over Bookmark this
0:24

THIS GUY CONNECTED HIS AI AGENTS TO HIS OBSIDIAN AND BUILT A BRAIN THAT LEARNS ON ITS OWN. HERE'S HOW TO BUILD IT Obsidian is just markdown files sitting in a folder. That turns out to be the perfect memory for an AI agent, because an agent can read and write those files directly. He wired his agents into the vault so they pull context from it, do the work, and write what they learned back. The notes aren't the point. The loop is, and it gets sharper every cycle How to build it: 1. Point an agent at your vault. The fastest way, no plugins, no API keys: open a terminal and run npx obsidian-mcp /path/to/your/vault. That exposes your Obsidian folder to Claude as a tool it can read, search, and write to. Add it to your Claude Code or Cowork config and restart 2. Confirm it can see the brain. Ask it: "list the notes in my vault and summarize what's in them." If it reads them back, the connection is live. Now it starts every task with everything the vault already holds instead of from zero 3. Give each agent one job and a write-back rule. Tell it: "research this, then save what you found as a new note in /brain with links to related notes." One agent researches, one summarizes, one plans. Each writes its output back into the vault 4. Close the loop. Add one line to every agent's instructions: "read /brain before starting, write your result back when done." Now each task leaves the vault richer, and the next run reads that before it works. It compounds instead of resetting 5. You only steer. Review what the brain produces, point it at the next thing. The agents handle the reading, writing, and connecting The edge isn't better notes. It's a brain that feeds itself, so the work gets sharper every cycle instead of starting over Bookmark this

Yarchi

58,186 просмотров • 2 месяцев назад

OXFORD AI PROFESSOR FED 100,000 MANUSCRIPTS DATING BACK TO 3000 BCE INTO A SINGLE KNOWLEDGE GRAPH - AND THEN ASKED IT HOW MUCH TRUTH IS IN MORMONISM the answer came back at 2% this is what an AI brain looks like when you've trained it on over 100,000 historical manuscripts going back to 3000 BCE every node in the graph is a concept and every connection shows how ideas relate across 5,000 years of human knowledge the system didn't just give an answer - it traced its reasoning across thousands of interconnected sources and cross-referenced historical records before producing a single number he's now taking questions on what to ask it next the list of topics people want tested is exactly what you'd expect.
0:27

OXFORD AI PROFESSOR FED 100,000 MANUSCRIPTS DATING BACK TO 3000 BCE INTO A SINGLE KNOWLEDGE GRAPH - AND THEN ASKED IT HOW MUCH TRUTH IS IN MORMONISM the answer came back at 2% this is what an AI brain looks like when you've trained it on over 100,000 historical manuscripts going back to 3000 BCE every node in the graph is a concept and every connection shows how ideas relate across 5,000 years of human knowledge the system didn't just give an answer - it traced its reasoning across thousands of interconnected sources and cross-referenced historical records before producing a single number he's now taking questions on what to ask it next the list of topics people want tested is exactly what you'd expect.

Noisy

267,677 просмотров • 2 месяцев назад

iit's wild that this still isn't the default way people use AI. Andrew Ng, a co-founder of Google Brain, gave a talk that crossed 2 million views: stop feeding your AI a folder of notes, build it a graph it can actually think with. his point is simple, stop making your AI read the whole vault to answer one thing. give it a graph, a router that points, an index that scores, edges that link, and it walks straight to the answer. it compounds, cheaper and faster the more you add. here's the whole thing: > point Claude at your notes and have it write a tiny router, your CLAUDE.md > build an index, one line per note, so it always knows where an answer lives > keep each note to one idea, and link the ones that depend on each other > make retrieval run as code, so it scores and opens the right files before the model thinks > tell it to prove the graph beats a flat folder on speed and cost, then keep the index fresh on every write set it up once and your AI stops guessing. it already knows your world before you type a word. full 12-step guide below. bookmark this.
0:23

iit's wild that this still isn't the default way people use AI. Andrew Ng, a co-founder of Google Brain, gave a talk that crossed 2 million views: stop feeding your AI a folder of notes, build it a graph it can actually think with. his point is simple, stop making your AI read the whole vault to answer one thing. give it a graph, a router that points, an index that scores, edges that link, and it walks straight to the answer. it compounds, cheaper and faster the more you add. here's the whole thing: > point Claude at your notes and have it write a tiny router, your CLAUDE.md > build an index, one line per note, so it always knows where an answer lives > keep each note to one idea, and link the ones that depend on each other > make retrieval run as code, so it scores and opens the right files before the model thinks > tell it to prove the graph beats a flat folder on speed and cost, then keep the index fresh on every write set it up once and your AI stops guessing. it already knows your world before you type a word. full 12-step guide below. bookmark this.

KingWilliam

15,099 просмотров • 1 месяц назад

I’m excited to announce a new course with DeepLearning.AI - Building Agentic RAG 💫 In this course, you’ll learn how to build a research assistant that can reason over multiple documents and answer complex questions. You’ll also learn how to step through the execution of the agent and steer it with human feedback. This represents a big step beyond any standard RAG pipeline, which is mostly good for simple questions over a small set of documents. Learn the layers first and then put them together: ✅ Routing ✅ Tool Use ✅ Multi-step reasoning with Memory ✅ Tool retrieval ✅ Debugging + user input Check it out!
2:06

I’m excited to announce a new course with DeepLearning.AI - Building Agentic RAG 💫 In this course, you’ll learn how to build a research assistant that can reason over multiple documents and answer complex questions. You’ll also learn how to step through the execution of the agent and steer it with human feedback. This represents a big step beyond any standard RAG pipeline, which is mostly good for simple questions over a small set of documents. Learn the layers first and then put them together: ✅ Routing ✅ Tool Use ✅ Multi-step reasoning with Memory ✅ Tool retrieval ✅ Debugging + user input Check it out!

Jerry Liu

76,293 просмотров • 2 лет назад

Geoffrey Hinton just made every AI critic accidentally describe their own brain. Hinton: “They shouldn’t be called hallucinations. They should be called confabulations.” One word. The entire debate unravels. The tech industry sees AI produce a confident wrong answer and calls it a defect. A bug to patch. They are measuring intelligence against the standard of a filing cabinet. And exposing that they understand neither. Hinton: “It’s not that there’s a file stored somewhere in your brain, like in a filing cabinet or in a computer memory.” Your brain does not store memories. It rebuilds them from nothing every time you remember. Fills gaps it never discloses. Fabricates details you would stake your life on. Then hands it all to you as truth. Hinton: “If I ask you to remember something that happened a few years ago, you’ll construct something that seems very plausible to you. And some of the details will be right and some will be wrong.” The wrong parts feel identical to the right ones. No internal warning. No distinction between what was remembered and what was invented on the spot. You have argued over memories that were partially fiction. Told stories about your own life that your brain manufactured in real time. With total conviction. And never once suspected. This is not a defect in human cognition. This IS cognition. The mechanism that fabricates is the same one that reasons, creates, and makes connections no one taught it to make. Not a separate system. Same architecture. Same process. You cannot remove the confabulation without killing the intelligence. They are the same thing. Hinton: “Psychologists have been studying confabulation in people since at least the 1930s.” A century of evidence. No one called the human brain broken. The moment a machine runs on the same principle, the world calls it defective. The people demanding AI that never gets a single detail wrong are not asking for intelligence. They’re asking for a search engine that sounds articulate. What we built is something else entirely. A system that thinks the way thinking actually works. Not retrieval. Construction. The imperfection is not the cost of intelligence. It is the signature.
1:05

Geoffrey Hinton just made every AI critic accidentally describe their own brain. Hinton: “They shouldn’t be called hallucinations. They should be called confabulations.” One word. The entire debate unravels. The tech industry sees AI produce a confident wrong answer and calls it a defect. A bug to patch. They are measuring intelligence against the standard of a filing cabinet. And exposing that they understand neither. Hinton: “It’s not that there’s a file stored somewhere in your brain, like in a filing cabinet or in a computer memory.” Your brain does not store memories. It rebuilds them from nothing every time you remember. Fills gaps it never discloses. Fabricates details you would stake your life on. Then hands it all to you as truth. Hinton: “If I ask you to remember something that happened a few years ago, you’ll construct something that seems very plausible to you. And some of the details will be right and some will be wrong.” The wrong parts feel identical to the right ones. No internal warning. No distinction between what was remembered and what was invented on the spot. You have argued over memories that were partially fiction. Told stories about your own life that your brain manufactured in real time. With total conviction. And never once suspected. This is not a defect in human cognition. This IS cognition. The mechanism that fabricates is the same one that reasons, creates, and makes connections no one taught it to make. Not a separate system. Same architecture. Same process. You cannot remove the confabulation without killing the intelligence. They are the same thing. Hinton: “Psychologists have been studying confabulation in people since at least the 1930s.” A century of evidence. No one called the human brain broken. The moment a machine runs on the same principle, the world calls it defective. The people demanding AI that never gets a single detail wrong are not asking for intelligence. They’re asking for a search engine that sounds articulate. What we built is something else entirely. A system that thinks the way thinking actually works. Not retrieval. Construction. The imperfection is not the cost of intelligence. It is the signature.

Dustin

16,260 просмотров • 1 месяц назад

ANTHROPIC KILLED THE VECTOR DATABASE AND REPLACED IT WITH A FOLDER no embeddings. no black-box retrieval. no special memory tools. the agent just reads and writes a folder the insight: your agent already knows how to use a filesystem. so Anthropic made long-term memory a directory of small markdown files, mounted straight into the agent's sandbox at /mnt/memory the agent manages it with the tools it already has, bash, read, write, edit, grep, and Anthropic auto-injects a note into the system prompt so the agent always knows the memory is there and where to look the result: > every change is an immutable version: full audit trail, point-in-time recovery, redaction > read-only or read-write per store, so many agents can safely share one memory > workspace-scoped and persistent, memory that survives every session end most teams bolt on a vector DB and pray retrieval works. Anthropic went the opposite way: compiled knowledge as plain, human-readable files you can open, diff, and edit yourself and there's a second layer. Anthropic's "Dreaming" runs offline, reviews past sessions and the store, and rewrites it into a cleaner, higher-signal version. the agent literally consolidates memory in its sleep. bookmark this. the most production-grade agent memory right now is just a well-organized folder, and Anthropic proved it.
0:50

ANTHROPIC KILLED THE VECTOR DATABASE AND REPLACED IT WITH A FOLDER no embeddings. no black-box retrieval. no special memory tools. the agent just reads and writes a folder the insight: your agent already knows how to use a filesystem. so Anthropic made long-term memory a directory of small markdown files, mounted straight into the agent's sandbox at /mnt/memory the agent manages it with the tools it already has, bash, read, write, edit, grep, and Anthropic auto-injects a note into the system prompt so the agent always knows the memory is there and where to look the result: > every change is an immutable version: full audit trail, point-in-time recovery, redaction > read-only or read-write per store, so many agents can safely share one memory > workspace-scoped and persistent, memory that survives every session end most teams bolt on a vector DB and pray retrieval works. Anthropic went the opposite way: compiled knowledge as plain, human-readable files you can open, diff, and edit yourself and there's a second layer. Anthropic's "Dreaming" runs offline, reviews past sessions and the store, and rewrites it into a cleaner, higher-signal version. the agent literally consolidates memory in its sleep. bookmark this. the most production-grade agent memory right now is just a well-organized folder, and Anthropic proved it.

NO1ennn

13,359 просмотров • 17 дней назад

MICROSOFT TRIED TO BUY IT. HARVARD AND THREE US UNIVERSITIES RELEASED IT FOR FREE a team from Harvard, Michigan State, Minnesota, and Georgia ran the first hard study on how agent memory behaves over time. the finding is uncomfortab they call it the "experience-following property": when a new task looks like a memory the agent pulls up, its answer copies that memory's answer almost 1:1 (correlation up to 0.95). the agent isn't reasoning - it's imitating that breaks in two ways: > error propagation - one wrong memory repeats on every similar task after it > misaligned replay - "successful" past runs turn out low-value or misleading when reused so "just add more memory" quietly makes agents worse. what fixed it: > store only high-quality experiences, not everything > use future task outcomes as free labels to delete low-value memories > pair selective adding with utility-based deletion the payoff: ~10% absolute accuracy gains, a cleaner memory bank, and agents stable under distribution shift no company owns this. they dropped the whole thing open-source, for free bookmark this. the next edge in AI agents isn't more memory - it's cleaner memory
0:23

MICROSOFT TRIED TO BUY IT. HARVARD AND THREE US UNIVERSITIES RELEASED IT FOR FREE a team from Harvard, Michigan State, Minnesota, and Georgia ran the first hard study on how agent memory behaves over time. the finding is uncomfortab they call it the "experience-following property": when a new task looks like a memory the agent pulls up, its answer copies that memory's answer almost 1:1 (correlation up to 0.95). the agent isn't reasoning - it's imitating that breaks in two ways: > error propagation - one wrong memory repeats on every similar task after it > misaligned replay - "successful" past runs turn out low-value or misleading when reused so "just add more memory" quietly makes agents worse. what fixed it: > store only high-quality experiences, not everything > use future task outcomes as free labels to delete low-value memories > pair selective adding with utility-based deletion the payoff: ~10% absolute accuracy gains, a cleaner memory bank, and agents stable under distribution shift no company owns this. they dropped the whole thing open-source, for free bookmark this. the next edge in AI agents isn't more memory - it's cleaner memory

NO1ennn

27,496 просмотров • 15 дней назад

Meet the minds building supermemory! 🧠 Our research engineer Prasanna breaks down how memory actually works, in your head, and in AI. inside the episode: → why forgetting is a feature and not a failure → what your hippocampus inside brain is really doing all day → how "dreaming" turns a messy day into long-term memory → what the AI labs get right, and wrong, when they copy the brain → why a bigger context window still isn't memory the line worth sitting with: stop asking where memory is stored. start asking what gets kept, and why the brain chose it. full episode below ↓
19:55

Meet the minds building supermemory! 🧠 Our research engineer Prasanna breaks down how memory actually works, in your head, and in AI. inside the episode: → why forgetting is a feature and not a failure → what your hippocampus inside brain is really doing all day → how "dreaming" turns a messy day into long-term memory → what the AI labs get right, and wrong, when they copy the brain → why a bigger context window still isn't memory the line worth sitting with: stop asking where memory is stored. start asking what gets kept, and why the brain chose it. full episode below ↓

supermemory

13,926 просмотров • 1 месяц назад

Introducing Memory 101 🧠 Every ai tool you use is a goldfish. brilliant for one chat, close the tab and it forgets you completely. memory is the most underbuilt layer in ai right now. so we're explaining how it actually works, one episode at a time. coming up: → rag vs memory, why most "ai memory" is just search → the 5 things real memory needs: facts, profiles, graphs, time, forgetting → why good memory has to forget things follow along!
0:50

Introducing Memory 101 🧠 Every ai tool you use is a goldfish. brilliant for one chat, close the tab and it forgets you completely. memory is the most underbuilt layer in ai right now. so we're explaining how it actually works, one episode at a time. coming up: → rag vs memory, why most "ai memory" is just search → the 5 things real memory needs: facts, profiles, graphs, time, forgetting → why good memory has to forget things follow along!

supermemory

17,235 просмотров • 2 месяцев назад

EINSTEIN’S BRAIN AND THE NEURONS THAT TAUGHT THEMSELVES PONG When Einstein died, scientists studied his brain to see what set it apart. Most of it looked ordinary, except for one thing: the folds. His brain had far more of them than average. More folds meant more surface area, and more surface area meant more neurons. The difference was concentrated in his parietal lobes, the regions tied to math and spatial reasoning. Genius left a trace, but it was subtle, hidden in structure rather than size. Decades later, another mystery arrived. In 2022, researchers at Cortical Labs grew human neurons in a Petri dish and wired them to a chip running the game Pong. There was no human player, no brain, and no mind, just cells, signals, and feedback. Yet somehow, the neurons learned. The longer they played, the better they became. That forced an uncomfortable question: how does a group of disconnected neurons know how to play a game it has never seen. The answer is still unclear. Einstein’s brain revealed how intelligence can be built. Source: Joe Rogan
0:48

EINSTEIN’S BRAIN AND THE NEURONS THAT TAUGHT THEMSELVES PONG When Einstein died, scientists studied his brain to see what set it apart. Most of it looked ordinary, except for one thing: the folds. His brain had far more of them than average. More folds meant more surface area, and more surface area meant more neurons. The difference was concentrated in his parietal lobes, the regions tied to math and spatial reasoning. Genius left a trace, but it was subtle, hidden in structure rather than size. Decades later, another mystery arrived. In 2022, researchers at Cortical Labs grew human neurons in a Petri dish and wired them to a chip running the game Pong. There was no human player, no brain, and no mind, just cells, signals, and feedback. Yet somehow, the neurons learned. The longer they played, the better they became. That forced an uncomfortable question: how does a group of disconnected neurons know how to play a game it has never seen. The answer is still unclear. Einstein’s brain revealed how intelligence can be built. Source: Joe Rogan

Mario Nawfal

86,839 просмотров • 8 месяцев назад

researchers gave a tiny local model human-style memory and its context limit basically stopped existing a team from MBZUAI, Princeton and Weizmann took a 1B model and rebuilt how it reads. instead of attending to everything at once, the model reads in 1,024 token chunks and passes the important stuff forward through an associative memory, the same way you carry the plot of a book between chapters without rereading them. the design mirrors human memory on purpose. full attention inside a chunk works as short-term memory. the module that carries information between chunks works as long-term memory. they even trained it like a person, starting with short easy texts and raising the difficulty gradually, because memory thrown into the deep end learns nothing. the numbers back it up. the normal model burns 40GB of GPU memory on a long document and collapses hard past its limit, dropping from 0.86 to 0.32 accuracy. the memory version holds 0.71 at double that length while using a flat 12GB no matter how long the input gets. it also needs about 30% fewer FLOPs. the part i keep thinking about is that nobody scaled anything here. they didn't build a bigger model, didn't stretch the window, didn't add compute. they looked at how a brain handles a long day and copied the architecture. a model small enough to run on a consumer gpu now survives documents its own architecture used to choke on. we keep treating intelligence as a compute problem. sometimes it's a memory problem.
0:43

researchers gave a tiny local model human-style memory and its context limit basically stopped existing a team from MBZUAI, Princeton and Weizmann took a 1B model and rebuilt how it reads. instead of attending to everything at once, the model reads in 1,024 token chunks and passes the important stuff forward through an associative memory, the same way you carry the plot of a book between chapters without rereading them. the design mirrors human memory on purpose. full attention inside a chunk works as short-term memory. the module that carries information between chunks works as long-term memory. they even trained it like a person, starting with short easy texts and raising the difficulty gradually, because memory thrown into the deep end learns nothing. the numbers back it up. the normal model burns 40GB of GPU memory on a long document and collapses hard past its limit, dropping from 0.86 to 0.32 accuracy. the memory version holds 0.71 at double that length while using a flat 12GB no matter how long the input gets. it also needs about 30% fewer FLOPs. the part i keep thinking about is that nobody scaled anything here. they didn't build a bigger model, didn't stretch the window, didn't add compute. they looked at how a brain handles a long day and copied the architecture. a model small enough to run on a consumer gpu now survives documents its own architecture used to choke on. we keep treating intelligence as a compute problem. sometimes it's a memory problem.

Alex Veremeyenko

16,147 просмотров • 1 месяц назад

Sam Altman, OpenAI CEO: "with an affordable amount of spend on tokens, you can do what a 100-person incredibly great engineering team would do as a startup." in this 40-minute Stanford talk, he breaks down the exact shift that lets a solo founder out-build a 100-person team. free. straight from the Stanford stage. worth more than a $2,000 startup accelerator. so i stopped arguing and built one. local agents. local tool. nothing leaving my laptop. an Obsidian clone, a knowledge graph, 13 agents a full second brain. solo. what a whole team would've shipped. minus the team. minus the cloud. here's the shift he breaks down and what i found building on it: - why tokens replaced headcount - the "one-person frontier lab" model ; your second brain is the lab - the inference layer every founder is sleeping on - why hiring 100 engineers is now the expensive way to build the 100-person team didn't get cheaper. it became one person, a second brain, and a swarm of agents. bookmark this. the full build is in the article below ⬇️
41:09

Sam Altman, OpenAI CEO: "with an affordable amount of spend on tokens, you can do what a 100-person incredibly great engineering team would do as a startup." in this 40-minute Stanford talk, he breaks down the exact shift that lets a solo founder out-build a 100-person team. free. straight from the Stanford stage. worth more than a $2,000 startup accelerator. so i stopped arguing and built one. local agents. local tool. nothing leaving my laptop. an Obsidian clone, a knowledge graph, 13 agents a full second brain. solo. what a whole team would've shipped. minus the team. minus the cloud. here's the shift he breaks down and what i found building on it: - why tokens replaced headcount - the "one-person frontier lab" model ; your second brain is the lab - the inference layer every founder is sleeping on - why hiring 100 engineers is now the expensive way to build the 100-person team didn't get cheaper. it became one person, a second brain, and a swarm of agents. bookmark this. the full build is in the article below ⬇️

Avid

65,153 просмотров • 2 месяцев назад

MICROSOFT QUIETLY SHIPPED THE RAG PIPELINE EVERYONE KEEPS TRYING TO BUILD FROM SCRATCH it is called Kernel Memory, and it turns any pile of files into an AI memory that answers with citations. drop your data in and Microsoft's pipeline does the rest: > ingest PDFs, Word, PPT, Excel, web pages, even images > extract the text, chunk it, embed it into vectors > store the vectors with metadata and keep the originals in blob storage > ask a question and it pulls the right chunks with semantic plus hybrid search then it hands the answer back grounded, with links to the exact source and full token usage. and it plugs into whatever you already run: > Azure OpenAI, OpenAI, Ollama, or Anthropic for the models > Qdrant, pgvector, Redis, or Azure AI Search for the vectors > every step is a swappable handler, so you can replace any part with your own one heads up: Microsoft archived the repo in June 2026 and ships it as a reference blueprint, not production. but as a map for how memory and RAG are actually built, nothing is cleaner. run it in one command: docker run kernelmemory/service bookmark this. Microsoft already drew the blueprint, you just build on it.
0:25

MICROSOFT QUIETLY SHIPPED THE RAG PIPELINE EVERYONE KEEPS TRYING TO BUILD FROM SCRATCH it is called Kernel Memory, and it turns any pile of files into an AI memory that answers with citations. drop your data in and Microsoft's pipeline does the rest: > ingest PDFs, Word, PPT, Excel, web pages, even images > extract the text, chunk it, embed it into vectors > store the vectors with metadata and keep the originals in blob storage > ask a question and it pulls the right chunks with semantic plus hybrid search then it hands the answer back grounded, with links to the exact source and full token usage. and it plugs into whatever you already run: > Azure OpenAI, OpenAI, Ollama, or Anthropic for the models > Qdrant, pgvector, Redis, or Azure AI Search for the vectors > every step is a swappable handler, so you can replace any part with your own one heads up: Microsoft archived the repo in June 2026 and ships it as a reference blueprint, not production. but as a map for how memory and RAG are actually built, nothing is cleaner. run it in one command: docker run kernelmemory/service bookmark this. Microsoft already drew the blueprint, you just build on it.

NO1ennn

82,462 просмотров • 5 дней назад

"I only like seeing Grok when, like, a bigot is trying to use Grok for something, and then Grok is like "this is the facts" and then they try to reword it in a way to get the answer they want". She's so awesome.
0:41

"I only like seeing Grok when, like, a bigot is trying to use Grok for something, and then Grok is like "this is the facts" and then they try to reword it in a way to get the answer they want". She's so awesome.

bint ♨️🐧

168,857 просмотров • 3 месяцев назад

Anthropic engineer just built a second brain that reads 3 files instead of 5,000, and cut the token bill 40% no orchestrator, no plugin, just a router file, an index, and code that scores before the model ever opens anything point claude at a flat folder of notes and every question reads all of them, that’s the default and it’s expensive this breakdown turns that exact pattern into a system anyone can run CLAUDE.md → index.md → nodes → edges → retrieval as code → memory → ship it twelve steps, one shared rule, evidence lives in the index, the model reads three files while the notes folder still holds thousands the stale index test, one line drifts and the whole graph quietly turns back into the pile it was built to escape, no exceptions save it and watch, then read the article below
0:25

Anthropic engineer just built a second brain that reads 3 files instead of 5,000, and cut the token bill 40% no orchestrator, no plugin, just a router file, an index, and code that scores before the model ever opens anything point claude at a flat folder of notes and every question reads all of them, that’s the default and it’s expensive this breakdown turns that exact pattern into a system anyone can run CLAUDE.md → index.md → nodes → edges → retrieval as code → memory → ship it twelve steps, one shared rule, evidence lives in the index, the model reads three files while the notes folder still holds thousands the stale index test, one line drifts and the whole graph quietly turns back into the pile it was built to escape, no exceptions save it and watch, then read the article below

leopardracer

37,304 просмотров • 28 дней назад

PSYCHO CREATED A GOD-MODE SECOND BRAIN ON FABLE 5 THAT FILLS ITS OWN KNOWLEDGE GAPS WHILE YOU SLEEP. KARPATHY'S FILE STRUCTURE MAKES IT POSSIBLE. Everyone wires Claude into Obsidian, posts the graph, and stops. this guy added a third system and it quietly changes the whole game: The full build is in the article below: vault structure, the Claude wiring, the exact prompts, the skills. > Obsidian as the knowledge base, but any notes app works as long as Claude Code can see the files > Karpathy's LLM Wiki doc as the skeleton, so the vault is organized the way a model actually reads, not the way humans decorate folders > a self-learning skill on a weekly loop: scans the entire brain, finds the gaps, then researches YouTube and Google by itself to fill them Read the third one again, actually your notes only ever contain what you happened to write down, and the gap-finder breaks that ceiling: the brain starts knowing things you never told it. Most second brains know what you told them. a real one notices what you didn't. that's the difference between note-taking and compounding. one sits there. the other gets smarter every week whether you showed up or not. P.S. send the article straight to your Claude, tell it to interview you, and the brain starts filling itself tonight.
1:05

PSYCHO CREATED A GOD-MODE SECOND BRAIN ON FABLE 5 THAT FILLS ITS OWN KNOWLEDGE GAPS WHILE YOU SLEEP. KARPATHY'S FILE STRUCTURE MAKES IT POSSIBLE. Everyone wires Claude into Obsidian, posts the graph, and stops. this guy added a third system and it quietly changes the whole game: The full build is in the article below: vault structure, the Claude wiring, the exact prompts, the skills. > Obsidian as the knowledge base, but any notes app works as long as Claude Code can see the files > Karpathy's LLM Wiki doc as the skeleton, so the vault is organized the way a model actually reads, not the way humans decorate folders > a self-learning skill on a weekly loop: scans the entire brain, finds the gaps, then researches YouTube and Google by itself to fill them Read the third one again, actually your notes only ever contain what you happened to write down, and the gap-finder breaks that ceiling: the brain starts knowing things you never told it. Most second brains know what you told them. a real one notices what you didn't. that's the difference between note-taking and compounding. one sits there. the other gets smarter every week whether you showed up or not. P.S. send the article straight to your Claude, tell it to interview you, and the brain starts filling itself tonight.

Miraqle

41,307 просмотров • 1 месяц назад

Introducing Cognee v1.0: a major breakthrough in agentic intelligence. It is 145% better than Opus 4.8 and GPT 5.5 at long context memory retrieval. Cognee allows a 100 BILLION token context window 100,000x more than Claude. It's: - 6.9x cheaper than GPT 5.5 and Opus 4.8 - Cold starts in 350ms & searches in 260ms Why this matters: Today agents forget important context, redo tasks, waste tokens, and slow down as workflows get more complex. Cognee solves this. It’s not a place to build agents. It connects to the agents you’ve already built, across any platform, and makes them significantly cheaper, faster, and more accurate. Here's how it works:
3:01

Introducing Cognee v1.0: a major breakthrough in agentic intelligence. It is 145% better than Opus 4.8 and GPT 5.5 at long context memory retrieval. Cognee allows a 100 BILLION token context window 100,000x more than Claude. It's: - 6.9x cheaper than GPT 5.5 and Opus 4.8 - Cold starts in 350ms & searches in 260ms Why this matters: Today agents forget important context, redo tasks, waste tokens, and slow down as workflows get more complex. Cognee solves this. It’s not a place to build agents. It connects to the agents you’ve already built, across any platform, and makes them significantly cheaper, faster, and more accurate. Here's how it works:

Vasilije

847,433 просмотров • 2 месяцев назад

This girl from IBM explained how Claude works better than most AI engineers could in 8 minutes > search used to match words, not meaning > then AI learned to understand intent > then RAG gave LLMs external memory - so they could read your documents > then agents arrived - now the AI decides what to search, when, and why the hardest part of AI isn't generating the answer it's knowing what to look at first full video above
8:27

This girl from IBM explained how Claude works better than most AI engineers could in 8 minutes > search used to match words, not meaning > then AI learned to understand intent > then RAG gave LLMs external memory - so they could read your documents > then agents arrived - now the AI decides what to search, when, and why the hardest part of AI isn't generating the answer it's knowing what to look at first full video above

Mr. Buzzoni

531,070 просмотров • 3 месяцев назад

Be Smart as Karpathy Andrej Karpathy with Teamily AI 🧠 Your Personal Knowledge Base: ✅ Built in One Chat. 📈 Compounded via Conversations. Karpathy’s insight is spot on ( It attracts 10 million views in a few days. The idea is simple: AI should build personal knowledge from everything you feed it, so it stops rediscovering things from scratch like a Retrieval-Augmented Generation (RAG). But here’s the reality — most people aren’t Stanford PhD-level geeks like Karpathy. For the rest of us, operating a hacky collection of scripts and tools (Obsidian Web Clipper, Marp, Dataview, etc.) as seen in Karpathy’s idea file is far too complex ( The Internet needs an intuitive product where a personal knowledge base is a persistent, compounding artifact — one that grows alongside the content you consume, the contexts you inhabit, and the questions you ask. Teamily AI ( is the answer. The conversation IS the knowledge base. It’s an AI-native messenger where AI teammates join your chats. They remember your past discussions, your preferences, and your team’s context — getting smarter the more you talk. No setup. No complicated workflows. Just text as you normally do. Whether you’re saving articles and videos, brainstorming at work, or collaborating with colleagues, your AI teammates are right there. They listen, remember, and help — not from scratch every time, but by building a personal knowledge graph of everything you’re involved in. In essence, your knowledge compounds automatically. ✨ The user experience is effortless. Whenever you need a well-organized view of your data, just ask the "Personal AI" at the top of the Teamily window: "Visualize my personal knowledge base" Want to customize the style or indexes? Just chat with it. You define how you manage your knowledge. Our co-founder Aiden has prepared a short video to show you just how easy it is. 📽️
2:41

Be Smart as Karpathy Andrej Karpathy with Teamily AI 🧠 Your Personal Knowledge Base: ✅ Built in One Chat. 📈 Compounded via Conversations. Karpathy’s insight is spot on ( It attracts 10 million views in a few days. The idea is simple: AI should build personal knowledge from everything you feed it, so it stops rediscovering things from scratch like a Retrieval-Augmented Generation (RAG). But here’s the reality — most people aren’t Stanford PhD-level geeks like Karpathy. For the rest of us, operating a hacky collection of scripts and tools (Obsidian Web Clipper, Marp, Dataview, etc.) as seen in Karpathy’s idea file is far too complex ( The Internet needs an intuitive product where a personal knowledge base is a persistent, compounding artifact — one that grows alongside the content you consume, the contexts you inhabit, and the questions you ask. Teamily AI ( is the answer. The conversation IS the knowledge base. It’s an AI-native messenger where AI teammates join your chats. They remember your past discussions, your preferences, and your team’s context — getting smarter the more you talk. No setup. No complicated workflows. Just text as you normally do. Whether you’re saving articles and videos, brainstorming at work, or collaborating with colleagues, your AI teammates are right there. They listen, remember, and help — not from scratch every time, but by building a personal knowledge graph of everything you’re involved in. In essence, your knowledge compounds automatically. ✨ The user experience is effortless. Whenever you need a well-organized view of your data, just ask the "Personal AI" at the top of the Teamily window: "Visualize my personal knowledge base" Want to customize the style or indexes? Just chat with it. You define how you manage your knowledge. Our co-founder Aiden has prepared a short video to show you just how easy it is. 📽️

Teamily AI

15,642 просмотров • 4 месяцев назад