Загрузка видео...

Не удалось загрузить видео

На главную

Part 1: CUDA Graphs - Intro & Basics - how normal CPU ←→ GPU execution works - what a kernel launch is and where CPU launch overhead comes from - why launching many small kernels individually can become a bottleneck - what CUDA Graphs are and how they reduce...

mohitwt_'s profile picture

mohit

7,686 subscribers

29,135 просмотров • 2 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 13

Фото профиля Manoj
Manoj2 месяцев назад

So if a beginner like me can follow u from day 1 Any prerequisites for this?

Фото профиля mohit
mohit2 месяцев назад

i would say you would still need some inference knowledge or context of if u get what I mean because I'm doing topics that I haven't deeply done or missing in upcoming days, so following my exact path as beginner won't work but if u happen to learn about cuda graphs or torch.compile, later ptx etc, you can easily watch and follow these topics

Фото профиля adi
adi2 месяцев назад

my goat

Фото профиля prasanna
prasanna2 месяцев назад

Bro one shottted ? Sick man keep it coming

Фото профиля mohit
mohit2 месяцев назад

nah lol, ~60 tries for 2 clips - its 2 seperate clips that i merged

Фото профиля Sisyphus
Sisyphus2 месяцев назад

Start a youtube

Фото профиля jbz
jbz2 месяцев назад

Please help me with DMing you.

Фото профиля mohit
mohit2 месяцев назад

sure

Фото профиля jbz
jbz2 месяцев назад

still can’t

Фото профиля mohit
mohit2 месяцев назад

dmed you

Фото профиля AI Apps API
AI Apps API2 месяцев назад

Great breakdown. Where this really bites is LLM inference decoding: you launch a flurry of tiny kernels per token and CPU launch overhead dominates at small batch sizes. Capturing the decode step as a graph and replaying it is one of the biggest easy wins for tokens per second.

Фото профиля Vishnu
Vishnu2 месяцев назад

🗿🗿

Фото профиля Vighnesh
Vighnesh2 месяцев назад

second follow

Похожие видео

Introducing The AI CUDA Engineer: An agentic AI system that automates the production of highly optimized CUDA kernels. The AI CUDA Engineer can produce highly optimized CUDA kernels, reaching 10-100x speedup over common machine learning operations in PyTorch. Our system is also able to produce highly optimized CUDA kernels that are much faster than existing CUDA kernels commonly used in production. We believe that fundamentally, AI systems can and should be as resource-efficient as the human brain, and that the best path to achieve this efficiency is to use AI to make AI more efficient! We are excited to publish our paper, The AI CUDA Engineer: Agentic CUDA Kernel Discovery, Optimization and Composition. We also release a dataset of over 17,000 verified CUDA kernels produced by The AI CUDA Engineer. Paper: Kernel Archive Webpage: HuggingFace Dataset: The AI CUDA Engineer utilizes evolutionary LLM-driven code optimization to autonomously improve the runtime of machine learning operations. Our system is not only able to convert PyTorch code into CUDA kernels, but through the use of evolution, it can also optimize the runtime performance of CUDA kernels, fuse multiple operations, and even discover novel solutions for writing efficient CUDA operations by learning from past innovations! We believe The AI CUDA Engineer opens a new era of AI-driven acceleration of AI and automated inference time optimization. We (Robert Lange, Aaditya Prasad 🇺🇸, sssss, Maxence Faldor, Yujin Tang, hardmaru) are excited to continue Sakana AI's mission of leveraging AI to improve AI.
1:21

Introducing The AI CUDA Engineer: An agentic AI system that automates the production of highly optimized CUDA kernels. The AI CUDA Engineer can produce highly optimized CUDA kernels, reaching 10-100x speedup over common machine learning operations in PyTorch. Our system is also able to produce highly optimized CUDA kernels that are much faster than existing CUDA kernels commonly used in production. We believe that fundamentally, AI systems can and should be as resource-efficient as the human brain, and that the best path to achieve this efficiency is to use AI to make AI more efficient! We are excited to publish our paper, The AI CUDA Engineer: Agentic CUDA Kernel Discovery, Optimization and Composition. We also release a dataset of over 17,000 verified CUDA kernels produced by The AI CUDA Engineer. Paper: Kernel Archive Webpage: HuggingFace Dataset: The AI CUDA Engineer utilizes evolutionary LLM-driven code optimization to autonomously improve the runtime of machine learning operations. Our system is not only able to convert PyTorch code into CUDA kernels, but through the use of evolution, it can also optimize the runtime performance of CUDA kernels, fuse multiple operations, and even discover novel solutions for writing efficient CUDA operations by learning from past innovations! We believe The AI CUDA Engineer opens a new era of AI-driven acceleration of AI and automated inference time optimization. We (Robert Lange, Aaditya Prasad 🇺🇸, sssss, Maxence Faldor, Yujin Tang, hardmaru) are excited to continue Sakana AI's mission of leveraging AI to improve AI.

Sakana AI

1,160,948 просмотров • 1 год назад

This is the best 1 hour on graph engineering ever recorded, just released by Google how to go from a single agent to a full 24/7 system from scratch: 00:00 - what graphs actually are 09:16 - your first working agent 21:15 - graph engineering explained 41:03 - graph engineering in practice 52:21 - graphs that improve themselves I've seen $500 courses that teach less than this video alone. Watch it, then take it further with my step-by-step guide on graph engineering below.
1:00:22

This is the best 1 hour on graph engineering ever recorded, just released by Google how to go from a single agent to a full 24/7 system from scratch: 00:00 - what graphs actually are 09:16 - your first working agent 21:15 - graph engineering explained 41:03 - graph engineering in practice 52:21 - graphs that improve themselves I've seen $500 courses that teach less than this video alone. Watch it, then take it further with my step-by-step guide on graph engineering below.

Anatoli Kopadze

122,845 просмотров • 1 месяц назад

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to an agent graph that builds itself: 10:16 - Build your first AI agent 41:05 - Master prompt engineering 54:45 - Turn agents into graphs 1:20:10 - Run loops inside agent graphs 1:43:33 - Build a graph that builds itself Most people ship one agent and stop there. Google is teaching everything after: Prompt → Agents → Graphs → Loops → Self-Building Systems Single agents are the old workflow. Graphs that evolve themselves are the new one. This 2-hour watch is worth more than most paid agent engineering courses. Bookmark and watch it today Then read how to run 1,000 agents from one prompt below ↓
1:54:59

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to an agent graph that builds itself: 10:16 - Build your first AI agent 41:05 - Master prompt engineering 54:45 - Turn agents into graphs 1:20:10 - Run loops inside agent graphs 1:43:33 - Build a graph that builds itself Most people ship one agent and stop there. Google is teaching everything after: Prompt → Agents → Graphs → Loops → Self-Building Systems Single agents are the old workflow. Graphs that evolve themselves are the new one. This 2-hour watch is worth more than most paid agent engineering courses. Bookmark and watch it today Then read how to run 1,000 agents from one prompt below ↓

Lunar

52,359 просмотров • 2 месяцев назад

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to an agent graph that can build itself: 0% → 10:16 - build your first AI agent 25% → 41:05 - master prompt engineering 50% → 54:45 - turn agents into graphs 75% → 1:20:10 - run loops inside agent graphs 100% → 1:43:33 - build a graph that builds itself Most people build one agent and stop there. Google is teaching everything that comes after: Prompt → Agents → Graphs → Loops → Self-Building Systems Single agents are the old workflow. Graphs that evolve themselves are the next one. This 2-hour course is worth more than most paid agent engineering courses. Bookmark it and watch today Then read how to run 1,000 agents from one prompt below ↓
1:55:41

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to an agent graph that can build itself: 0% → 10:16 - build your first AI agent 25% → 41:05 - master prompt engineering 50% → 54:45 - turn agents into graphs 75% → 1:20:10 - run loops inside agent graphs 100% → 1:43:33 - build a graph that builds itself Most people build one agent and stop there. Google is teaching everything that comes after: Prompt → Agents → Graphs → Loops → Self-Building Systems Single agents are the old workflow. Graphs that evolve themselves are the next one. This 2-hour course is worth more than most paid agent engineering courses. Bookmark it and watch today Then read how to run 1,000 agents from one prompt below ↓

Lunar

44,889 просмотров • 1 месяц назад

Our new short course, Knowledge Graphs for RAG, is now available! Knowledge graphs are a data structure that is great at capturing complex relationships between data of multiple types. By enabling more sophisticated retrieval of text than similarity search alone, knowledge graphs can improve the context you pass to the LLM and the performance of your RAG applications. In this course, taught by Andreas Kollegger of Neo4j, you’ll - Explore how knowledge graphs work by building a graph of public financial documents from scratch - Learn to write queries that retrieve text and data from the graph and use it to enhance the context you pass to an LLM chatbot - Combine a knowledge graph with a question-answer chain to build better RAG-powered chat systems Sign up here!
3:46

Our new short course, Knowledge Graphs for RAG, is now available! Knowledge graphs are a data structure that is great at capturing complex relationships between data of multiple types. By enabling more sophisticated retrieval of text than similarity search alone, knowledge graphs can improve the context you pass to the LLM and the performance of your RAG applications. In this course, taught by Andreas Kollegger of Neo4j, you’ll - Explore how knowledge graphs work by building a graph of public financial documents from scratch - Learn to write queries that retrieve text and data from the graph and use it to enhance the context you pass to an LLM chatbot - Combine a knowledge graph with a question-answer chain to build better RAG-powered chat systems Sign up here!

Andrew Ng

244,435 просмотров • 2 лет назад

IBM just turned five of its best graph lectures into a 50-minute course on building AI systems around connected knowledge: • 00:00 - How knowledge graphs represent entities and relationships • 05:36 - Why vector-based RAG misses connections between facts • 09:53 - GraphRAG, precision retrieval and context engineering • 20:43 - How graph neural networks learn from connected data • 37:30 - Building GraphRAG with knowledge graphs and Cypher The progression is simple: Text chunks → entities → relationships → graph retrieval → graph intelligence Vector search finds information that looks similar. Graphs recover the structure behind that information. This 50-minute course will teach you more about modern graph engineering than most paid AI programs. Watch it today, then read how to become a knowledge graph engineer in the article below.
48:59

IBM just turned five of its best graph lectures into a 50-minute course on building AI systems around connected knowledge: • 00:00 - How knowledge graphs represent entities and relationships • 05:36 - Why vector-based RAG misses connections between facts • 09:53 - GraphRAG, precision retrieval and context engineering • 20:43 - How graph neural networks learn from connected data • 37:30 - Building GraphRAG with knowledge graphs and Cypher The progression is simple: Text chunks → entities → relationships → graph retrieval → graph intelligence Vector search finds information that looks similar. Graphs recover the structure behind that information. This 50-minute course will teach you more about modern graph engineering than most paid AI programs. Watch it today, then read how to become a knowledge graph engineer in the article below.

Morlex

12,053 просмотров • 2 месяцев назад

Build better RAG by letting a team of agents extract and connect your reference materials into a knowledge graph. Our new short course, “Agentic Knowledge Graph Construction,” taught by Neo4j Innovation Lead Andreas Kollegger, shows you how. Knowledge graphs are an important way to store information accurately but they are a lot of work to build manually. In this course you’ll learn how to build a team of agents that turn data– in this case product reviews and invoices from suppliers–into structured graphs of entities and relationships for RAG. Learn how agents can automatically handle the time-consuming work of building graphs — extracting entities and relationships (e.g., Product "contains" Assembly, Part "supplied_by" Supplier, Customer review "mentions" Product), deduplicating them, fact-checking them, and committing them to a graph database — so your retrieval system can find right information to generate accurate output. For example, you can use agents to help trace customer complaints directly to specific suppliers, manufacturing processes, and product hierarchies, thus turning fragmented information into queryable business intelligence. Skills you’ll gain: - Build, store, and access knowledge graphs using the Neo4j graph database - Build multi-agent systems using Google’s Agent Development Kit (ADK) - Set up a loop of agentic workflows to propose and refine a graph schema through fact-checking - Connect agent-generated graphs of unstructured and structured data into a unified knowledge graph This course gets into the practicum of why knowledge graphs give more accurate information retrieval than vector search alone, especially for high-stakes applications where precision matters more than fuzzy similarity matching. Sign up here:
2:26

Build better RAG by letting a team of agents extract and connect your reference materials into a knowledge graph. Our new short course, “Agentic Knowledge Graph Construction,” taught by Neo4j Innovation Lead Andreas Kollegger, shows you how. Knowledge graphs are an important way to store information accurately but they are a lot of work to build manually. In this course you’ll learn how to build a team of agents that turn data– in this case product reviews and invoices from suppliers–into structured graphs of entities and relationships for RAG. Learn how agents can automatically handle the time-consuming work of building graphs — extracting entities and relationships (e.g., Product "contains" Assembly, Part "supplied_by" Supplier, Customer review "mentions" Product), deduplicating them, fact-checking them, and committing them to a graph database — so your retrieval system can find right information to generate accurate output. For example, you can use agents to help trace customer complaints directly to specific suppliers, manufacturing processes, and product hierarchies, thus turning fragmented information into queryable business intelligence. Skills you’ll gain: - Build, store, and access knowledge graphs using the Neo4j graph database - Build multi-agent systems using Google’s Agent Development Kit (ADK) - Set up a loop of agentic workflows to propose and refine a graph schema through fact-checking - Connect agent-generated graphs of unstructured and structured data into a unified knowledge graph This course gets into the practicum of why knowledge graphs give more accurate information retrieval than vector search alone, especially for high-stakes applications where precision matters more than fuzzy similarity matching. Sign up here:

Andrew Ng

168,153 просмотров • 1 год назад

Andrew Ng just released a 1-hour course on what comes after RAG. How to turn raw documents into agent memory: 00:00 - Why larger context windows are not enough 02:18 - Extract entities, relationships, and sources 14:31 - Coordinate specialized agents inside one graph 30:43 - Build it with Google ADK and Neo4j 47:25 - Turn knowledge graphs into persistent memory Most people are still learning RAG. Andrew Ng is already teaching the layer that comes after it. Documents → Entities → Relationships → Agents → Knowledge Graph More context is the old workflow. Persistent graphs are the new memory layer. This 1-hour watch is worth more than most paid GraphRAG courses. Bookmark and watch it today Then read the full graph engineering roadmap below
1:00:53

Andrew Ng just released a 1-hour course on what comes after RAG. How to turn raw documents into agent memory: 00:00 - Why larger context windows are not enough 02:18 - Extract entities, relationships, and sources 14:31 - Coordinate specialized agents inside one graph 30:43 - Build it with Google ADK and Neo4j 47:25 - Turn knowledge graphs into persistent memory Most people are still learning RAG. Andrew Ng is already teaching the layer that comes after it. Documents → Entities → Relationships → Agents → Knowledge Graph More context is the old workflow. Persistent graphs are the new memory layer. This 1-hour watch is worth more than most paid GraphRAG courses. Bookmark and watch it today Then read the full graph engineering roadmap below

Lunar

42,118 просмотров • 2 месяцев назад

I added a Knowledge Graph to Cursor using MCP. You gotta see this working! Knowledge graphs are a game-changer for AI Agents, and this is one example of how you can take advantage of them. How this works: 1. Cursor connects to Graphiti's MCP Server. Graphiti is a very popular open-source Knowledge Graph library for AI agents. 2. Graphiti connects to Neo4j running locally. Now, every time I interact with Cursor, the information is synthesized and stored in the knowledge graph. In short, Cursor now "remembers" everything about our project. Huge! Here is the video I recorded.
11:39

I added a Knowledge Graph to Cursor using MCP. You gotta see this working! Knowledge graphs are a game-changer for AI Agents, and this is one example of how you can take advantage of them. How this works: 1. Cursor connects to Graphiti's MCP Server. Graphiti is a very popular open-source Knowledge Graph library for AI agents. 2. Graphiti connects to Neo4j running locally. Now, every time I interact with Cursor, the information is synthesized and stored in the knowledge graph. In short, Cursor now "remembers" everything about our project. Huge! Here is the video I recorded.

Santiago

155,647 просмотров • 1 год назад

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to 100 agents running inside one graph: 17:44 - Build your first AI agent 39:30 - Run agents with loop engineering 1:12:38 - Turn agent loops into graphs 1:34:26 - Build agents that throttle themselves 1:55:05 - Orchestrate the full multi-agent system Most people build one agent and stop there. Google is teaching the full stack: Prompt → Agents → Loops → Graphs → Multi-Agent Systems Single agents are the old workflow. Self-regulating agent graphs are the new one. This 2-hour watch is worth more than most paid agent engineering courses. Bookmark and watch it today Then read the full architecture below ↓
2:05:53

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to 100 agents running inside one graph: 17:44 - Build your first AI agent 39:30 - Run agents with loop engineering 1:12:38 - Turn agent loops into graphs 1:34:26 - Build agents that throttle themselves 1:55:05 - Orchestrate the full multi-agent system Most people build one agent and stop there. Google is teaching the full stack: Prompt → Agents → Loops → Graphs → Multi-Agent Systems Single agents are the old workflow. Self-regulating agent graphs are the new one. This 2-hour watch is worth more than most paid agent engineering courses. Bookmark and watch it today Then read the full architecture below ↓

Lunar

403,731 просмотров • 2 месяцев назад

Knowledge graphs for representing information are unbeatable. After this, you will never build a RAG system without knowledge graphs. It will take you five lines of code to build a knowledge graph with your data. I recorded a video to show you how you can do this. I used Cognee, an open-source library that outperforms any basic vector search approach in terms of retrieval relevance. They are collaborating with me on this post. Cognee is: • Easy to use • Reduces hallucinations • Open-source Here is a link to the repository: They also offer a comprehensive platform and UI with Python notebooks you can utilize to manage your data. Here is the link:
6:02

Knowledge graphs for representing information are unbeatable. After this, you will never build a RAG system without knowledge graphs. It will take you five lines of code to build a knowledge graph with your data. I recorded a video to show you how you can do this. I used Cognee, an open-source library that outperforms any basic vector search approach in terms of retrieval relevance. They are collaborating with me on this post. Cognee is: • Easy to use • Reduces hallucinations • Open-source Here is a link to the repository: They also offer a comprehensive platform and UI with Python notebooks you can utilize to manage your data. Here is the link:

Santiago

126,565 просмотров • 1 год назад

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to 100 agents running inside a single graph: 0% → 17:44 - build your first AI agent 25% → 39:30 - run agents with loop engineering 50% → 1:12:38 - turn agent loops into graphs 75% → 1:34:26 - build agents that regulate themselves 100% → 1:55:05 - orchestrate the entire multi-agent system Most people build one agent and stop there. Google is teaching the entire stack: Prompt → Agents → Loops → Graphs → Multi-Agent Systems Single agents are the old workflow. Self-regulating agent graphs are the next one. This 2-hour course is worth more than most paid agent engineering courses. Bookmark it and watch today Then read the full architecture below ↓
2:04:23

Google just released a free 2-hour course on full Graph Engineering. How to go from one prompt to 100 agents running inside a single graph: 0% → 17:44 - build your first AI agent 25% → 39:30 - run agents with loop engineering 50% → 1:12:38 - turn agent loops into graphs 75% → 1:34:26 - build agents that regulate themselves 100% → 1:55:05 - orchestrate the entire multi-agent system Most people build one agent and stop there. Google is teaching the entire stack: Prompt → Agents → Loops → Graphs → Multi-Agent Systems Single agents are the old workflow. Self-regulating agent graphs are the next one. This 2-hour course is worth more than most paid agent engineering courses. Bookmark it and watch today Then read the full architecture below ↓

Lunar

12,509 просмотров • 1 месяц назад

Ben Thompson explains how LLMs greatly diminished Nvidia's CUDA moat even as they sent the stock to the moon "So the weird thing about large language models is they were obviously incredible for Nvidia. That's why their stock went to the moon." "They have been on and off the most valuable company in the world." "It was also very bad for Nvidia. And the reason it was bad for Nvidia is that the play with CUDA is to build a developer ecosystem on top of CUDA." "But CUDA only works on Nvidia GPUs. So you get CUDA for free. It's easier to use, and it's a tremendous investment. Nvidia almost went under trying to build CUDA at a time when no one understood what they were doing or why they were wasting money on it." "And that's why Jensen Huang will get bristly, particularly when people question their rent-seeking or profit, whatever. It's like, no, they earned their spot fair and square." "Absolutely. It shouldn't be forgotten. They have earned every dollar they've gotten through 25 years of taking massive risks." "It bottomed out in October 2022. I wrote an article like three weeks before ChatGPT came out, tracing their bottoming-out history and their search for what was next." "'Nvidia in the Valley.' So, go back to this GTC. So I wrote an article at the time called 'Nvidia Waves and Moats'." "And what was interesting about that GTC was, number one, it was very boring. All the cool stuff kind of got scrubbed out." "Now, Jensen Huang has brought that stuff back, so the last few GTCs he's more talking about other things. Now it comes across as, oh, you're still looking for something beyond the LLM." "Because the problem with the LLM is it shifts the developer platform far above where Nvidia sits. All the activity is happening on top of LLMs. And so no one who's writing an AI application today is using CUDA." "Now, some people are, if you're training your own model and you're doing some low-level things or non-LLM things." "But the vast majority of the energy and all the money and the ecosystem is far removed from CUDA." "They have no idea and don't need to know or care what chips their application is running on. They're just on the OpenAI API, or the Anthropic API, or using Bedrock on Amazon, and it's sitting on Trainium, and they're using a Chinese open-source model. It's totally abstracted away, and this is why LLMs were bad for Nvidia." "Now, again, all the money they made along the way is worth it, but their moat has been tremendously diminished." "CUDA is still a moat if you need to do stuff that requires CUDA. But the vast majority of stuff, in energy, doesn't require CUDA, like in a post-LLM world."
3:04

Ben Thompson explains how LLMs greatly diminished Nvidia's CUDA moat even as they sent the stock to the moon "So the weird thing about large language models is they were obviously incredible for Nvidia. That's why their stock went to the moon." "They have been on and off the most valuable company in the world." "It was also very bad for Nvidia. And the reason it was bad for Nvidia is that the play with CUDA is to build a developer ecosystem on top of CUDA." "But CUDA only works on Nvidia GPUs. So you get CUDA for free. It's easier to use, and it's a tremendous investment. Nvidia almost went under trying to build CUDA at a time when no one understood what they were doing or why they were wasting money on it." "And that's why Jensen Huang will get bristly, particularly when people question their rent-seeking or profit, whatever. It's like, no, they earned their spot fair and square." "Absolutely. It shouldn't be forgotten. They have earned every dollar they've gotten through 25 years of taking massive risks." "It bottomed out in October 2022. I wrote an article like three weeks before ChatGPT came out, tracing their bottoming-out history and their search for what was next." "'Nvidia in the Valley.' So, go back to this GTC. So I wrote an article at the time called 'Nvidia Waves and Moats'." "And what was interesting about that GTC was, number one, it was very boring. All the cool stuff kind of got scrubbed out." "Now, Jensen Huang has brought that stuff back, so the last few GTCs he's more talking about other things. Now it comes across as, oh, you're still looking for something beyond the LLM." "Because the problem with the LLM is it shifts the developer platform far above where Nvidia sits. All the activity is happening on top of LLMs. And so no one who's writing an AI application today is using CUDA." "Now, some people are, if you're training your own model and you're doing some low-level things or non-LLM things." "But the vast majority of the energy and all the money and the ecosystem is far removed from CUDA." "They have no idea and don't need to know or care what chips their application is running on. They're just on the OpenAI API, or the Anthropic API, or using Bedrock on Amazon, and it's sitting on Trainium, and they're using a Chinese open-source model. It's totally abstracted away, and this is why LLMs were bad for Nvidia." "Now, again, all the money they made along the way is worth it, but their moat has been tremendously diminished." "CUDA is still a moat if you need to do stuff that requires CUDA. But the vast majority of stuff, in energy, doesn't require CUDA, like in a post-LLM world."

Fireside Alpha

12,980 просмотров • 1 месяц назад

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

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

Hanako

73,867 просмотров • 1 месяц назад

Andrew Ng just released a free 2-hour course on complete Graph Engineering How to go from one prompt to 100 agents that loop, improve, and run without you: 09:14 - Build your first AI agent 33:11 - Run agents with loop engineering 1:02:46 - Turn agent loops into graphs 1:30:15 - Build agents that rewrite themselves 1:49:05 - Run the full graph system without you Most people are still building one agent and calling it finished Andrew Ng is already teaching everything that comes next: Prompt → Agents → Loops → Graphs → Self-Improving Systems Single agents are the old workflow Systems that improve and run without you are the next one This 2-hour course is worth more than most $500 agent engineering programs Bookmark it and watch before everyone catches up Then read how to run 1,000 agents from one prompt below ↓
1:51:29

Andrew Ng just released a free 2-hour course on complete Graph Engineering How to go from one prompt to 100 agents that loop, improve, and run without you: 09:14 - Build your first AI agent 33:11 - Run agents with loop engineering 1:02:46 - Turn agent loops into graphs 1:30:15 - Build agents that rewrite themselves 1:49:05 - Run the full graph system without you Most people are still building one agent and calling it finished Andrew Ng is already teaching everything that comes next: Prompt → Agents → Loops → Graphs → Self-Improving Systems Single agents are the old workflow Systems that improve and run without you are the next one This 2-hour course is worth more than most $500 agent engineering programs Bookmark it and watch before everyone catches up Then read how to run 1,000 agents from one prompt below ↓

reira

14,899 просмотров • 1 месяц назад

Chamath: US AI Startups Can Learn A Lot from DeepSeek "This is a case where necessity was the mother of invention." On E213, Chamath Palihapitiya explained how, even if the $6M number is inaccurate, DeepSeek still had some impressive breakthroughs: GRPO > PPO for reinforcement learning "These guys were like, 'Well, how am I going to do this reinforcement learning thing?' They invented a totally different algorithm." "It uses a lot less computer memory and it's highly performant." PTX > CUDA to build the model "And then the second thing that was crazy is everybody is used to building models and compiling through CUDA, which is NVIDIA's proprietary language." "And these guys worked totally around CUDA and they did something called PTX, which goes right to the bare metal." What US AI startups can learn from this: "We, meaning the West, with all the money that we've had, didn't come up with these ideas." "And I think part of why we didn't come up is not that we're not smart enough to do it, but we weren't forced to because the constraints didn't exist." "And so I just wonder how we make sure we learn this principle, meaning, when the AI company wakes up and rolls out of bed and some VC gives them $200M, maybe that's not the right answer for a Series A or a Seed." "And maybe the right answer is $2M so that they do these DeepSeek-like innovations."
0:55

Chamath: US AI Startups Can Learn A Lot from DeepSeek "This is a case where necessity was the mother of invention." On E213, Chamath Palihapitiya explained how, even if the $6M number is inaccurate, DeepSeek still had some impressive breakthroughs: GRPO > PPO for reinforcement learning "These guys were like, 'Well, how am I going to do this reinforcement learning thing?' They invented a totally different algorithm." "It uses a lot less computer memory and it's highly performant." PTX > CUDA to build the model "And then the second thing that was crazy is everybody is used to building models and compiling through CUDA, which is NVIDIA's proprietary language." "And these guys worked totally around CUDA and they did something called PTX, which goes right to the bare metal." What US AI startups can learn from this: "We, meaning the West, with all the money that we've had, didn't come up with these ideas." "And I think part of why we didn't come up is not that we're not smart enough to do it, but we weren't forced to because the constraints didn't exist." "And so I just wonder how we make sure we learn this principle, meaning, when the AI company wakes up and rolls out of bed and some VC gives them $200M, maybe that's not the right answer for a Series A or a Seed." "And maybe the right answer is $2M so that they do these DeepSeek-like innovations."

The All-In Podcast

108,815 просмотров • 1 год назад

Congrats to the Kimi.ai team! This is awesome. Great to see this level of research coming from open-source frontier model labs. I liked the paper so much I built a Rust implementation of it ;) Full AttnRes + Block AttnRes with two-phase inference, built using Burn (tensor library and Deep Learning Framework, in Rust, by Tracel AI). Runs on CPU, CUDA, Metal, wgpu. Includes an interactive TUI that trains a model live and visualizes depth attention evolving from uniform to selective in real time. Repo link and more on what is implemented in the comments.
0:43

Congrats to the Kimi.ai team! This is awesome. Great to see this level of research coming from open-source frontier model labs. I liked the paper so much I built a Rust implementation of it ;) Full AttnRes + Block AttnRes with two-phase inference, built using Burn (tensor library and Deep Learning Framework, in Rust, by Tracel AI). Runs on CPU, CUDA, Metal, wgpu. Includes an interactive TUI that trains a model live and visualizes depth attention evolving from uniform to selective in real time. Repo link and more on what is implemented in the comments.

abdel

95,651 просмотров • 6 месяцев назад

Andrew Ng just released a 2-hour course on full Graph Engineering. How to go from one prompt to 100 agents that loop, rewrite themselves, and run without you: 09:14 - Build your first AI agent 33:11 - Run agents with loop engineering 1:02:46 - Turn agent loops into graphs 1:30:15 - Build agents that rewrite themselves 1:49:05 - Run the full graph system without you Most people are still building one agent and calling it done. Andrew Ng is already teaching everything after: Prompt → Agents → Loops → Graphs → Self-Improving Systems Single agents are the old workflow. Systems that improve and run without you are the new one. This 2-hour watch is worth more than most $500 agent engineering courses. Bookmark and watch it before everyone catches up. Then read how to run 1,000 agents from one prompt below ↓
1:48:23

Andrew Ng just released a 2-hour course on full Graph Engineering. How to go from one prompt to 100 agents that loop, rewrite themselves, and run without you: 09:14 - Build your first AI agent 33:11 - Run agents with loop engineering 1:02:46 - Turn agent loops into graphs 1:30:15 - Build agents that rewrite themselves 1:49:05 - Run the full graph system without you Most people are still building one agent and calling it done. Andrew Ng is already teaching everything after: Prompt → Agents → Loops → Graphs → Self-Improving Systems Single agents are the old workflow. Systems that improve and run without you are the new one. This 2-hour watch is worth more than most $500 agent engineering courses. Bookmark and watch it before everyone catches up. Then read how to run 1,000 agents from one prompt below ↓

Lunar

226,799 просмотров • 1 месяц назад

Google just released a free 2-hour+ course on full agent engineering. How to go from one prompt to a system that runs while you sleep: 38:46 - Build your first AI agent 54:46 - Connect MCP tools 1:12:43 - Run agents with four loop patterns 1:20:57 - Turn agent loops into graphs 2:22:31 - Build the complete autonomous system Most people are still building one agent and stopping there. Google is already teaching the full stack: Agents → Tools → Loops → Graphs → Autonomous Systems Single agents are the old workflow. Systems that run without you are the new one. This course is worth more than most paid agent engineering bootcamps. Bookmark and watch it today Then read the full graph engineering playbook below
2:14:04

Google just released a free 2-hour+ course on full agent engineering. How to go from one prompt to a system that runs while you sleep: 38:46 - Build your first AI agent 54:46 - Connect MCP tools 1:12:43 - Run agents with four loop patterns 1:20:57 - Turn agent loops into graphs 2:22:31 - Build the complete autonomous system Most people are still building one agent and stopping there. Google is already teaching the full stack: Agents → Tools → Loops → Graphs → Autonomous Systems Single agents are the old workflow. Systems that run without you are the new one. This course is worth more than most paid agent engineering bootcamps. Bookmark and watch it today Then read the full graph engineering playbook below

Lunar

198,150 просмотров • 2 месяцев назад

Context vs. Graphs, clearly explained! context engineering is great, but it has a ceiling: you can make one window perfect. there is still only one of it. every technique on that layer is rationing the same scarce thing. compact, retrieve less, delete, defer. all of it is deciding what to drop. Graph engineering fixes this by moving the decision up a layer: not what goes in the window, but how many windows there are and what each one is for. you need both. here's how it works: ↳ inside a window: context engineering. what loads, in what order, what gets compacted ↳ between windows: the graph. how many lanes, what each one is allowed to see, what comes back Prompts → Context → Harness → Loops → Graphs each lane gets a clean window, nothing in one competes with anything in another, and your main thread stops filling up. the trick is being selective about what comes back. a subagent reads six thousand tokens of files and hands you a four hundred token summary. that ratio is the whole point. send back the raw material instead and you have moved the problem, not solved it. one thing to know before you scale it. not everything survives compaction equally, and almost nobody knows the table. ↳ the project-root rules file and auto memory are re-injected from disk. they come back intact ↳ path-scoped rules and nested rules files live in message history. they get summarized away and do not return until a matching file is read again so a rule that genuinely must persist cannot be path-scoped. move it to the root and pay the always-loaded cost, or accept that it is advisory in any long session. and the one that eats whole nights: shared context makes parallel agents converge. four auditors on one window produce one opinion with three echoes. you paid four times for it. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this and read it below ↓
0:53

Context vs. Graphs, clearly explained! context engineering is great, but it has a ceiling: you can make one window perfect. there is still only one of it. every technique on that layer is rationing the same scarce thing. compact, retrieve less, delete, defer. all of it is deciding what to drop. Graph engineering fixes this by moving the decision up a layer: not what goes in the window, but how many windows there are and what each one is for. you need both. here's how it works: ↳ inside a window: context engineering. what loads, in what order, what gets compacted ↳ between windows: the graph. how many lanes, what each one is allowed to see, what comes back Prompts → Context → Harness → Loops → Graphs each lane gets a clean window, nothing in one competes with anything in another, and your main thread stops filling up. the trick is being selective about what comes back. a subagent reads six thousand tokens of files and hands you a four hundred token summary. that ratio is the whole point. send back the raw material instead and you have moved the problem, not solved it. one thing to know before you scale it. not everything survives compaction equally, and almost nobody knows the table. ↳ the project-root rules file and auto memory are re-injected from disk. they come back intact ↳ path-scoped rules and nested rules files live in message history. they get summarized away and do not return until a matching file is read again so a rule that genuinely must persist cannot be path-scoped. move it to the root and pay the always-loaded cost, or accept that it is advisory in any long session. and the one that eats whole nights: shared context makes parallel agents converge. four auditors on one window produce one opinion with three echoes. you paid four times for it. 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

36,970 просмотров • 1 месяц назад