Загрузка видео...

Не удалось загрузить видео

На главную

sharing a pretty important primitive in my agentic engineering setup I call it "gnhf" - good night, have fun basically, every night before I go to bed, I would put my agents to work so I never wake up "empty-handed". it's done through a similar setup as geoff's famous...

kunchenguid's profile picture

Kun Chen

13,070 subscribers

59,814 просмотров • 5 месяцев 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

Loop engineering is great until something breaks. Here is how I improve the reliability of my agentic loops. I use human-in-the-loop (HITL). It's easy and extremely effective. Anyone can build this. My setup: I recorded a quick demo of how it all works. I shared recently that I now use more voice agents to build and communicate with agents. I also use them to verify. I hate the idea of being tied down to my computer or in a Slack channel to communicate with my agents. Here is what I have done to streamline communication with my agents. All my Claude and Codex agent sessions now use the Dial MCP server. It has a bunch of tools and provisions my agents with their own number that can place calls as native tools, with voice, SMS, and iMessage behind one interface. As my loops/automations work on PRs and new features, my agents escalate decisions to me via a short phone call. This is extremely useful when I am on the road or away from my desk. If you want to try this with Claude Code or Codex, paste this into your agent and get started right away: "Get yourself a Dial phone number and call me. Say hello and that setup is working, then hang up. Follow Natan Voitenkov and team are building something special here. Go check them out. Give your agent a phone number now: ($5 free credit)
5:02

Loop engineering is great until something breaks. Here is how I improve the reliability of my agentic loops. I use human-in-the-loop (HITL). It's easy and extremely effective. Anyone can build this. My setup: I recorded a quick demo of how it all works. I shared recently that I now use more voice agents to build and communicate with agents. I also use them to verify. I hate the idea of being tied down to my computer or in a Slack channel to communicate with my agents. Here is what I have done to streamline communication with my agents. All my Claude and Codex agent sessions now use the Dial MCP server. It has a bunch of tools and provisions my agents with their own number that can place calls as native tools, with voice, SMS, and iMessage behind one interface. As my loops/automations work on PRs and new features, my agents escalate decisions to me via a short phone call. This is extremely useful when I am on the road or away from my desk. If you want to try this with Claude Code or Codex, paste this into your agent and get started right away: "Get yourself a Dial phone number and call me. Say hello and that setup is working, then hang up. Follow Natan Voitenkov and team are building something special here. Go check them out. Give your agent a phone number now: ($5 free credit)

elvis

15,295 просмотров • 2 месяцев назад

Three skills I use every day in Claude Code and Codex to solve my hardest problems: 1️⃣ /agent-watchdog When I have one agent like Codex working on a task and I don't fully trust it's going to do everything right, I'll open up another one like Claude Code and tell it to watchdog the Codex thread. You can copy the Codex deep link into Claude Code and it'll look at the prompt you sent, watch the Codex thread until it's done, then compare the Codex solution to how it was planning to solve it and automatically fix anything that Codex missed. It can also test the work of the other agent end-to-end. Similar to the idea of OpenRouter's new Fusion feature, I've definitely found that two models thinking through a problem and checking each other's work can be wildly more impactful than just one. 2️⃣ /plan-arbiter Similar ideas as /agent-watchdog - but with this one you have both make plans, compare plans, negotiate the differences, and make a final plan to execute. I find Claude Code is better at writing plans, but Codex is faster and cheaper to execute on them. Then I usually have Claude Code watchdog the Codex work and fix anything that was missed. 3️⃣ /read-the-damn-docs One thing that drives me crazy with coding agents is they're so reluctant to look up docs. They'll just guess and guess and guess at the right API surface for things, or the right solution to an integration of two things. Once I explicitly tell it to look up the docs, it says "Oh, I see the answer," and it fixes the problem. So I made the /read-the-damn-docs skill. Add it and your agents will know when and how to do efficient web searches to look up docs for the types of problems you really should look up docs for. All of these are totally open source over on my GitHub. If you try them, let me know your feedback. Will link to them below:
1:21

Three skills I use every day in Claude Code and Codex to solve my hardest problems: 1️⃣ /agent-watchdog When I have one agent like Codex working on a task and I don't fully trust it's going to do everything right, I'll open up another one like Claude Code and tell it to watchdog the Codex thread. You can copy the Codex deep link into Claude Code and it'll look at the prompt you sent, watch the Codex thread until it's done, then compare the Codex solution to how it was planning to solve it and automatically fix anything that Codex missed. It can also test the work of the other agent end-to-end. Similar to the idea of OpenRouter's new Fusion feature, I've definitely found that two models thinking through a problem and checking each other's work can be wildly more impactful than just one. 2️⃣ /plan-arbiter Similar ideas as /agent-watchdog - but with this one you have both make plans, compare plans, negotiate the differences, and make a final plan to execute. I find Claude Code is better at writing plans, but Codex is faster and cheaper to execute on them. Then I usually have Claude Code watchdog the Codex work and fix anything that was missed. 3️⃣ /read-the-damn-docs One thing that drives me crazy with coding agents is they're so reluctant to look up docs. They'll just guess and guess and guess at the right API surface for things, or the right solution to an integration of two things. Once I explicitly tell it to look up the docs, it says "Oh, I see the answer," and it fixes the problem. So I made the /read-the-damn-docs skill. Add it and your agents will know when and how to do efficient web searches to look up docs for the types of problems you really should look up docs for. All of these are totally open source over on my GitHub. If you try them, let me know your feedback. Will link to them below:

Steve (Builder.io)

43,089 просмотров • 2 месяцев назад

Finally got a chance to play around with Andrej Karpathy's LLM Council. I built it as a plugin inside of Claude Code. Hooked it up with OpenRouter for models. The AskUserQuestion tool came in handy to select the council and chairman. This is my first test, but I agree with Karpathy that the concept of LLM ensembles can be used beyond models that offer perspectives on interesting questions. I feel like this could have really cool applications in agentic coding. More on that soon. I built this as a plugin, so next I will be exploring other user cases around agentic coding, like evaluation, tool building, designing, and research. If there is enough interest, I will clean it up and push it out as an open plugin.
0:38

Finally got a chance to play around with Andrej Karpathy's LLM Council. I built it as a plugin inside of Claude Code. Hooked it up with OpenRouter for models. The AskUserQuestion tool came in handy to select the council and chairman. This is my first test, but I agree with Karpathy that the concept of LLM ensembles can be used beyond models that offer perspectives on interesting questions. I feel like this could have really cool applications in agentic coding. More on that soon. I built this as a plugin, so next I will be exploring other user cases around agentic coding, like evaluation, tool building, designing, and research. If there is enough interest, I will clean it up and push it out as an open plugin.

elvis

79,795 просмотров • 8 месяцев назад

I made a dynamic waves system for Clear Water 3! It's pretty simple, but fun to play with. You can see the waves and flowmaps at work too. I'm debating over how to package up these systems that require extra setup, but I plan on including it. #vrchat #unity
0:26

I made a dynamic waves system for Clear Water 3! It's pretty simple, but fun to play with. You can see the waves and flowmaps at work too. I'm debating over how to package up these systems that require extra setup, but I plan on including it. #vrchat #unity

Silent

46,742 просмотров • 2 месяцев назад

Tried to make it as simple as possible. Follow it and give it a try. There are more things you can do but i kept it easy for everyone for now. Check the first comment for the link. I have the tool and some documentation for further reading. Good luck.
5:06

Tried to make it as simple as possible. Follow it and give it a try. There are more things you can do but i kept it easy for everyone for now. Check the first comment for the link. I have the tool and some documentation for further reading. Good luck.

elias

90,000 просмотров • 2 лет назад

Jensen Huang reveals how he actually spends his time as CEO and why he often refuses to sleep "I spend a lot of time talking to our employees. People are surprised how much time I spend eating in the cafeteria, whether it's lunch or dinner" "I need to come back to my number one job function and what I believe is the most important responsibility I have, which is to be the custodian of the culture" "If you want to be the custodian of the culture, you can't do it through CNN, you can't do it through Forbes magazine articles" "You have to do it 1% at a time unfortunately, or large crowds at a time" "I try to spend my time as much as I can on the things that I believe will have long-lasting influence on our company" "As a CEO, your time is not always yours. You need to have the discipline to make it yours. I'll often come into the office and tell my admin to clear my calendars just so I can have that time back" "You also come to the conclusion that as a CEO, not sleeping is a good choice. That's always a good option. It creates more time when you don't sleep"
4:26

Jensen Huang reveals how he actually spends his time as CEO and why he often refuses to sleep "I spend a lot of time talking to our employees. People are surprised how much time I spend eating in the cafeteria, whether it's lunch or dinner" "I need to come back to my number one job function and what I believe is the most important responsibility I have, which is to be the custodian of the culture" "If you want to be the custodian of the culture, you can't do it through CNN, you can't do it through Forbes magazine articles" "You have to do it 1% at a time unfortunately, or large crowds at a time" "I try to spend my time as much as I can on the things that I believe will have long-lasting influence on our company" "As a CEO, your time is not always yours. You need to have the discipline to make it yours. I'll often come into the office and tell my admin to clear my calendars just so I can have that time back" "You also come to the conclusion that as a CEO, not sleeping is a good choice. That's always a good option. It creates more time when you don't sleep"

Jaynit

57,759 просмотров • 3 месяцев назад

🎙️: “this could be championship weekend for you lando norris” lando: “i'm excited for all of it. i've never been in this situation before but i'm excited, i think it's exciting for everyone. i think i certainly would have liked to have the 30 point gap instead of 24 to whoever it is, but yeah makes it more exciting for you guys i guess so good for you. and no, i think i'll just go and try and do the same thing i'll focus on trying to get good quali and have a good race and just take one step at a time. i don't mind, of course the earlier i can do it the better but at the same time that doesn't change how i do anything i just need to try and win and simple as that”
0:53

🎙️: “this could be championship weekend for you lando norris” lando: “i'm excited for all of it. i've never been in this situation before but i'm excited, i think it's exciting for everyone. i think i certainly would have liked to have the 30 point gap instead of 24 to whoever it is, but yeah makes it more exciting for you guys i guess so good for you. and no, i think i'll just go and try and do the same thing i'll focus on trying to get good quali and have a good race and just take one step at a time. i don't mind, of course the earlier i can do it the better but at the same time that doesn't change how i do anything i just need to try and win and simple as that”

ray

435,630 просмотров • 9 месяцев назад

Well, I *was* having fun with my first 2011 🤣 Tonight after a couple mags it failed to go fully into battery, so I pushed it forward and fired the chambered round (perhaps that was an error, though it was a pretty gentle push I didn't need to force it or anything). Now it's totally locked up, not sure what to do but maybe one of you knows. It only has about 300 rounds through it, and cleaned and oiled it after the first 100. I know it has a bit of a break-in period, though this was my first issue with it. But it's a pretty big issue! I could try and really force it open but I dont want to break it.
0:23

Well, I *was* having fun with my first 2011 🤣 Tonight after a couple mags it failed to go fully into battery, so I pushed it forward and fired the chambered round (perhaps that was an error, though it was a pretty gentle push I didn't need to force it or anything). Now it's totally locked up, not sure what to do but maybe one of you knows. It only has about 300 rounds through it, and cleaned and oiled it after the first 100. I know it has a bit of a break-in period, though this was my first issue with it. But it's a pretty big issue! I could try and really force it open but I dont want to break it.

Kostas Moros

60,014 просмотров • 5 месяцев назад

Y’all know I love to have a good time… but when it comes to betting, it's more fun when it's for fun. So keep it smart and play responsibly. DraftKings has tools to help you do just that! #DKPartner
0:32

Y’all know I love to have a good time… but when it comes to betting, it's more fun when it's for fun. So keep it smart and play responsibly. DraftKings has tools to help you do just that! #DKPartner

Kevin Hart

95,967 просмотров • 1 год назад

New Andrej Karpathy interview Says AI agent failures stem from user skill, not model capability. Poor instructions cause errors. He suggests delegating 20-minute macro actions like coding and research to parallel agents and reviewing their work. --- "I think everything, like so many things, even if they don't work, I think to a large extent you feel like it's a skill issue. It's not that the capability is not there; it's that you just haven't found a way to string together what's available. Like, I didn't give good enough instructions to the agents in the file, or whatever it may be. I don't have a nice enough memory tool that I put in there, or something like that. So, it all kind of feels like a skill issue when it doesn't work to some extent. You want to see how you can parallelize them, and you want to be a 'Pierce tender,' basically. Pierce famously has a funny photo where he's in front of lots of these Codex agents behind the monitor. They all take about 20 minutes if you run them correctly and use high effort. You have multiple—you know, 10 or 20—pull requests checked out. It's just like you can do much larger macro actions. It's not just, 'Here's a line of code, here's a new function.' It's like, 'Here's a new functionality, delegate it to agent one. Here's a new functionality that's not going to interfere with the other one, give it to agent two.' Then, you try to review their work as best as you can, depending on how much you care about that code. You look for these macro actions that you can manipulate your software repository by. Another agent is doing some research, another agent is writing code, another one is coming up with a plan for some new implementation. Everything just happens in these macro actions over your repository. You're just trying to become really good at it and develop a muscle memory for it. It's very rewarding when it actually works, but it's also a new thing to learn. Hence, the psychosis." --- From No Priors YT channel (link in comment)
2:05

New Andrej Karpathy interview Says AI agent failures stem from user skill, not model capability. Poor instructions cause errors. He suggests delegating 20-minute macro actions like coding and research to parallel agents and reviewing their work. --- "I think everything, like so many things, even if they don't work, I think to a large extent you feel like it's a skill issue. It's not that the capability is not there; it's that you just haven't found a way to string together what's available. Like, I didn't give good enough instructions to the agents in the file, or whatever it may be. I don't have a nice enough memory tool that I put in there, or something like that. So, it all kind of feels like a skill issue when it doesn't work to some extent. You want to see how you can parallelize them, and you want to be a 'Pierce tender,' basically. Pierce famously has a funny photo where he's in front of lots of these Codex agents behind the monitor. They all take about 20 minutes if you run them correctly and use high effort. You have multiple—you know, 10 or 20—pull requests checked out. It's just like you can do much larger macro actions. It's not just, 'Here's a line of code, here's a new function.' It's like, 'Here's a new functionality, delegate it to agent one. Here's a new functionality that's not going to interfere with the other one, give it to agent two.' Then, you try to review their work as best as you can, depending on how much you care about that code. You look for these macro actions that you can manipulate your software repository by. Another agent is doing some research, another agent is writing code, another one is coming up with a plan for some new implementation. Everything just happens in these macro actions over your repository. You're just trying to become really good at it and develop a muscle memory for it. It's very rewarding when it actually works, but it's also a new thing to learn. Hence, the psychosis." --- From No Priors YT channel (link in comment)

Rohan Paul

23,207 просмотров • 5 месяцев назад

I’m trying to compute the last couple days and honestly I’m absolutely fucking shattered. You have been my North Star for everything for as long as I can remember from when I was misunderstood as a child to the way people thought I was just a little “too much” or “strange” in my life and career. I owe so much to you, your wife and your family - you all gave me a road to run down and supported me when people would turn their nose up. You loved life so much and you adored music. It was your smile man. You taught me it was beautiful to be out of the box. I promise you with all my heart I will try my best and make it my life’s journey to keep the sprit that you started and what you have taught me alive. I will give it my best shot. To sing this song for you at an event surrounded by legends inspired by you to a crowd that loved you was truly my life’s greatest honour and I vow to play this song every night for the rest of my life. You’re my hero in every regard. I hope you’re up there avin a drink with Randy. For the prince of darkness you sure brought all the light to the world. I love you Ozzy.
0:59

I’m trying to compute the last couple days and honestly I’m absolutely fucking shattered. You have been my North Star for everything for as long as I can remember from when I was misunderstood as a child to the way people thought I was just a little “too much” or “strange” in my life and career. I owe so much to you, your wife and your family - you all gave me a road to run down and supported me when people would turn their nose up. You loved life so much and you adored music. It was your smile man. You taught me it was beautiful to be out of the box. I promise you with all my heart I will try my best and make it my life’s journey to keep the sprit that you started and what you have taught me alive. I will give it my best shot. To sing this song for you at an event surrounded by legends inspired by you to a crowd that loved you was truly my life’s greatest honour and I vow to play this song every night for the rest of my life. You’re my hero in every regard. I hope you’re up there avin a drink with Randy. For the prince of darkness you sure brought all the light to the world. I love you Ozzy.

YUNGBLUD

1,078,459 просмотров • 1 год назад

18 minutes of #KimSeonho yapping with Seonhohadas with his endless gratefulness during the fan sign event!! 🥹🥹 "i'm releasing many projects this year as you know. because i wasn't able to greet my fans through my work for a while, it feels like everything is being released at once this year, i really hope you'll be happy about that. I filmed them with a hope to give back a bit more, so i hope you'll enjoy and feel happy about it. I hope you have fun. I hope you feel my heart even if it's just through my work. i will also do my best in my work. I'll make sure to show you my best self."
17:46

18 minutes of #KimSeonho yapping with Seonhohadas with his endless gratefulness during the fan sign event!! 🥹🥹 "i'm releasing many projects this year as you know. because i wasn't able to greet my fans through my work for a while, it feels like everything is being released at once this year, i really hope you'll be happy about that. I filmed them with a hope to give back a bit more, so i hope you'll enjoy and feel happy about it. I hope you have fun. I hope you feel my heart even if it's just through my work. i will also do my best in my work. I'll make sure to show you my best self."

joy ◡̈ 🥐

367,963 просмотров • 7 месяцев назад

Bryson DeChambeau says rivalries are good for the game of golf and he’d love to go up against Rory McIlroy this week at the Ryder Cup: “Look, I think rivalries are good gore the game of golf, and albeit I have the ultimate respect for Rory as a player. It's going to be fun to go up against him this week, whether it's against him directly or through other players. I think it going to be a fun challenge this week. “Would I love to go up against him? Yeah. It would be a lot of fun. Is it going to happen? It's not likely. I mean, maybe once. You never know. I don't know if there's planning behind the scenes or whatnot. But look, he's a fierce competitor, a great competitor, but one that I would love the opportunity to play against this week.” Who would like to see these two go head to head and who do you think would win?
0:42

Bryson DeChambeau says rivalries are good for the game of golf and he’d love to go up against Rory McIlroy this week at the Ryder Cup: “Look, I think rivalries are good gore the game of golf, and albeit I have the ultimate respect for Rory as a player. It's going to be fun to go up against him this week, whether it's against him directly or through other players. I think it going to be a fun challenge this week. “Would I love to go up against him? Yeah. It would be a lot of fun. Is it going to happen? It's not likely. I mean, maybe once. You never know. I don't know if there's planning behind the scenes or whatnot. But look, he's a fierce competitor, a great competitor, but one that I would love the opportunity to play against this week.” Who would like to see these two go head to head and who do you think would win?

Flushing It

437,112 просмотров • 11 месяцев назад

You should try svelte... seriously I been using it for the last couple weeks after deciding to finally give it a try (Ben Davis kept bugging me)... now I can't go back... I love it... it's fun to write and the models are pretty good at it
9:43

You should try svelte... seriously I been using it for the last couple weeks after deciding to finally give it a try (Ben Davis kept bugging me)... now I can't go back... I love it... it's fun to write and the models are pretty good at it

Micky

67,245 просмотров • 5 месяцев назад

Just sorta looking at the leftovers. I have to live in a hotel for now....It's rough and stressful....But I'm not deterred. A lot of you guys asked me to set up a go fund me.... so I have. It's not a lot but it helps. Thank you all so much! Link in comments 🔗⬇️
1:18

Just sorta looking at the leftovers. I have to live in a hotel for now....It's rough and stressful....But I'm not deterred. A lot of you guys asked me to set up a go fund me.... so I have. It's not a lot but it helps. Thank you all so much! Link in comments 🔗⬇️

𝙍𝙀𝙔𝘾𝘼𝙇𝙇 _ 🎧🥬

444,438 просмотров • 2 лет назад

I finally got around to dogfooding my own ntm tool for managing a bunch of agents, and... it's not bad? I made an ntm skill for Claude Code and then simply told it to use ntm with 10 CCs and 5 Codex instances on a new project that I just finished making the beads for. Easy.😎
14:54

I finally got around to dogfooding my own ntm tool for managing a bunch of agents, and... it's not bad? I made an ntm skill for Claude Code and then simply told it to use ntm with 10 CCs and 5 Codex instances on a new project that I just finished making the beads for. Easy.😎

Jeffrey Emanuel

17,578 просмотров • 7 месяцев назад

Grok 3 was just released. You won't believe it, I've already created a game. (I got early access THIS MORNING). This game was 100% created by GROK, I just told it what I wanted, and put the code in the right place. I just keep asking for adjustments, and it keeps spitting the game out in a single file that I can put on my desktop and run. The game is changed forever. I've been developing a lot with AI's from every other major AI builder lately, trying to decide what I like best, and grok is a PLAYER. I don't have official benchmarks, and I don't have API setup yet so it's not my normal workflow, but it felt every bit as capable as Sonet, 4o, or anything else. In the next day or so I'll get it set up as part of my workflow in NVIM and put it to real work. This is incredible. We live in the future. Everyone is a developer now.
0:48

Grok 3 was just released. You won't believe it, I've already created a game. (I got early access THIS MORNING). This game was 100% created by GROK, I just told it what I wanted, and put the code in the right place. I just keep asking for adjustments, and it keeps spitting the game out in a single file that I can put on my desktop and run. The game is changed forever. I've been developing a lot with AI's from every other major AI builder lately, trying to decide what I like best, and grok is a PLAYER. I don't have official benchmarks, and I don't have API setup yet so it's not my normal workflow, but it felt every bit as capable as Sonet, 4o, or anything else. In the next day or so I'll get it set up as part of my workflow in NVIM and put it to real work. This is incredible. We live in the future. Everyone is a developer now.

Penny2x

1,606,293 просмотров • 1 год назад