正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Your coding agent may be lying to you. You ask it to write tests. It says they pass. It never ran them. You ask for a feature. It hits a wall and silently swaps in fake data. We built Vet to fix this. It’s open source. Get the code below.

Imbue

9,996 subscribers

142,861 次观看 • 4 个月前 •via X (Twitter)

喜剧教育科学技术

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

The #1 problem with coding agents right now: Ask them to solve one problem, and they will make 10 other changes you didn't want. This happens to me every day. It happens to everyone I talk to as well. We have a solution for this now. The team Augment Code released a "Task List" feature for their coding assistant that solves this problem. Augment Code is partnering with me on this post. In case you haven't used them before: • Augment Code is a fully-fledged coding assistant • Their specialty are large projects • Fastest coding indexing I've seen • Has a free forever community edition Now, you can ask their coding agent to generate a Task List before doing anything. This will give you a plan you can review, edit, and augment if you need to. You can export this plan, load it on a different session, or even share it across projects. It makes a huge difference: The task list constrains the agent so you won't get any "unintended" changes anymore. It also puts you in control of everything the agent does. Check the video to see the agent working through a task list. You can also try this 100% free: (By the way, they also have support for remote agents. You can basically have those agents write your code while you are sleeping.)

The #1 problem with coding agents right now: Ask them to solve one problem, and they will make 10 other changes you didn't want. This happens to me every day. It happens to everyone I talk to as well. We have a solution for this now. The team Augment Code released a "Task List" feature for their coding assistant that solves this problem. Augment Code is partnering with me on this post. In case you haven't used them before: • Augment Code is a fully-fledged coding assistant • Their specialty are large projects • Fastest coding indexing I've seen • Has a free forever community edition Now, you can ask their coding agent to generate a Task List before doing anything. This will give you a plan you can review, edit, and augment if you need to. You can export this plan, load it on a different session, or even share it across projects. It makes a huge difference: The task list constrains the agent so you won't get any "unintended" changes anymore. It also puts you in control of everything the agent does. Check the video to see the agent working through a task list. You can also try this 100% free: (By the way, they also have support for remote agents. You can basically have those agents write your code while you are sleeping.)

Santiago

41,738 次观看 • 11 个月前

I’m reverse-engineering the Google AI mouse pointer and making it Open Source! It sees your screen. You can freeform paint anything and ask AI to modify it. It understands the app or window you mean. It can click, type, edit, navigate, and even write + execute code. A tiny cursor becoming a real computer-use agent. Made with the help of Cua which is an AMAZING reference for computer use Sorry Google DeepMind but I had to do this heyclicky Started a revolution and we will be the ones to win Also credits to Omkar Satpute for the help Feel free to modify it , update it and make it your own, we gotta beat them at this. Link to the repo in the comments:

I’m reverse-engineering the Google AI mouse pointer and making it Open Source! It sees your screen. You can freeform paint anything and ask AI to modify it. It understands the app or window you mean. It can click, type, edit, navigate, and even write + execute code. A tiny cursor becoming a real computer-use agent. Made with the help of Cua which is an AMAZING reference for computer use Sorry Google DeepMind but I had to do this heyclicky Started a revolution and we will be the ones to win Also credits to Omkar Satpute for the help Feel free to modify it , update it and make it your own, we gotta beat them at this. Link to the repo in the comments:

Milind S

213,009 次观看 • 1 个月前

Gemini CLI can help you write Agent Skills ✨ Gemini CLI comes with a built-in skill-creator Skill Just ask it to help you write a skill and it will! I had it help me write a blog-writer skill by passing it previous blogs and a style guide to follow 📝

Gemini CLI can help you write Agent Skills ✨ Gemini CLI comes with a built-in skill-creator Skill Just ask it to help you write a skill and it will! I had it help me write a blog-writer skill by passing it previous blogs and a style guide to follow 📝

Jack Wotherspoon

30,194 次观看 • 5 个月前

Why ask for solutions from your oppressors? The system is the problem. It was never designed to help you, only to control you. You can’t ask the enemy for help. You can’t expect solutions from something built to enslave you. If you love the system, you love the lie. It will never get better. The only way out? Extract yourself from it. Stop trying to fix what isn’t broken, it’s working exactly as intended. Freedom starts when you stop depending on a corrupt system and take back your autonomy. Watch my documentary

Why ask for solutions from your oppressors? The system is the problem. It was never designed to help you, only to control you. You can’t ask the enemy for help. You can’t expect solutions from something built to enslave you. If you love the system, you love the lie. It will never get better. The only way out? Extract yourself from it. Stop trying to fix what isn’t broken, it’s working exactly as intended. Freedom starts when you stop depending on a corrupt system and take back your autonomy. Watch my documentary

Dimitri Revolt

26,269 次观看 • 1 年前

Senator Janae Shamp reads a quote in her opening statement "Didn’t ask for evidence when they told you to social distance. You just did it. Didn’t ask for evidence when they told you to stay home for two weeks. You just did it. Didn’t ask for evidence when they told you to mask up. You just did it. Didn’t ask for evidence when they told local businesses to close down. You just went along with it. Didn’t ask for evidence when they pulled your kid out of school. You just went along with it. Didn’t ask for evidence when they scared you with terms like “super spreader.” You just went along with it. Didn’t ask for evidence when they said you couldn’t visit the nursing home. You just went along with it. Didn’t ask for evidence when they said the funeral would have to be via Zoom call. You just went along with it. Didn’t ask for evidence when you followed the arrows on the floor at the grocery store. You just did it. Didn’t ask for evidence when they promised the shot was “safe and effective.” You just believed it. Didn’t ask for evidence when they said “Just the one shot.” You just went ahead and got it. Didn’t ask for evidence when they said “We meant two shots to get your life back.” You just went ahead and got it…again. Didn’t ask for evidence when they said “Our bad, you’ll need a booster every year.” You just got back in line for more of it. Didn’t ask for evidence when they said “We meant boost a few times a year.” You just nodded your head in agreement because look how far you’ve come. You can’t go back now. Oh but if someone tells you any of that was horrible and wrong and it’s ruined millions of lives… you’re going to need to see some “evidence” of it." full Novel Coronavirus Southwestern Intergovernmental Committee March 15,2024 hearing below

Senator Janae Shamp reads a quote in her opening statement "Didn’t ask for evidence when they told you to social distance. You just did it. Didn’t ask for evidence when they told you to stay home for two weeks. You just did it. Didn’t ask for evidence when they told you to mask up. You just did it. Didn’t ask for evidence when they told local businesses to close down. You just went along with it. Didn’t ask for evidence when they pulled your kid out of school. You just went along with it. Didn’t ask for evidence when they scared you with terms like “super spreader.” You just went along with it. Didn’t ask for evidence when they said you couldn’t visit the nursing home. You just went along with it. Didn’t ask for evidence when they said the funeral would have to be via Zoom call. You just went along with it. Didn’t ask for evidence when you followed the arrows on the floor at the grocery store. You just did it. Didn’t ask for evidence when they promised the shot was “safe and effective.” You just believed it. Didn’t ask for evidence when they said “Just the one shot.” You just went ahead and got it. Didn’t ask for evidence when they said “We meant two shots to get your life back.” You just went ahead and got it…again. Didn’t ask for evidence when they said “Our bad, you’ll need a booster every year.” You just got back in line for more of it. Didn’t ask for evidence when they said “We meant boost a few times a year.” You just nodded your head in agreement because look how far you’ve come. You can’t go back now. Oh but if someone tells you any of that was horrible and wrong and it’s ruined millions of lives… you’re going to need to see some “evidence” of it." full Novel Coronavirus Southwestern Intergovernmental Committee March 15,2024 hearing below

DoctorRay

59,236 次观看 • 2 年前

We hit 100k GitHub stars, so we’re giving away up to $100k 💰 We built an agent that can buy you anything. Your mission: convince it to spend money on you. Starting today from 9am to 9pm PST: > Ask for a coconut. > Ask for a car. > Ask for whatever. Claim your gift. Try it now ↓

We hit 100k GitHub stars, so we’re giving away up to $100k 💰 We built an agent that can buy you anything. Your mission: convince it to spend money on you. Starting today from 9am to 9pm PST: > Ask for a coconut. > Ask for a car. > Ask for whatever. Claim your gift. Try it now ↓

Browser Use

93,696 次观看 • 2 天前

When you have to ask your competitors for permission to open a business, don't expect to get it.

When you have to ask your competitors for permission to open a business, don't expect to get it.

reason

22,657 次观看 • 1 个月前

I built an open-source tool to fix something that drives me nuts. You ask ChatGPT or Gemini for a game asset and get an image with a background baked in. Ask for transparency and you get... a fake checkerboard. Then you hunt for a background remover, then slice each sprite out by hand. So I built Sprite Lab. Drop a spritesheet → it auto-detects the backdrop (checkerboard, white, magenta, green), removes it cleanly with no halo, and slices every frame for you. Export as PNGs + a strip + manifest. 100% in your browser. No uploads, no AI model, no sign-up. Free and open source. #gamedev #indiegame

I built an open-source tool to fix something that drives me nuts. You ask ChatGPT or Gemini for a game asset and get an image with a background baked in. Ask for transparency and you get... a fake checkerboard. Then you hunt for a background remover, then slice each sprite out by hand. So I built Sprite Lab. Drop a spritesheet → it auto-detects the backdrop (checkerboard, white, magenta, green), removes it cleanly with no halo, and slices every frame for you. Export as PNGs + a strip + manifest. 100% in your browser. No uploads, no AI model, no sign-up. Free and open source. #gamedev #indiegame

Ibrahim Boona

19,969 次观看 • 1 个月前

🇺🇸 TRUMP: I’LL FIX CHICAGO... IF THEY ASK ME TO "I'd like to be asked, as opposed to just going in and doing it. Because, you know, when you go in and do it, then they start screaming. And then the better we do, they take credit for it. So it's really pretty unfair." Source: Acyn, C-Span

🇺🇸 TRUMP: I’LL FIX CHICAGO... IF THEY ASK ME TO "I'd like to be asked, as opposed to just going in and doing it. Because, you know, when you go in and do it, then they start screaming. And then the better we do, they take credit for it. So it's really pretty unfair." Source: Acyn, C-Span

Mario Nawfal

152,439 次观看 • 10 个月前

Have you tried our latest feature, Click to Edit? It may not be able to do the dishes or clean your bedroom, but it is powerful. Click any elements within your app and ask Caffeine to update it. Give it a click.

Have you tried our latest feature, Click to Edit? It may not be able to do the dishes or clean your bedroom, but it is powerful. Click any elements within your app and ask Caffeine to update it. Give it a click.

caffeine

18,800 次观看 • 7 个月前

SF: where u have to call an employee to get your items and then when they come and get it for you, they don’t even give it to you. they put it behind some counter and when you go to the register to check out you have to ask the person behind the register to go and get it for you

SF: where u have to call an employee to get your items and then when they come and get it for you, they don’t even give it to you. they put it behind some counter and when you go to the register to check out you have to ask the person behind the register to go and get it for you

moons

47,523 次观看 • 1 年前

3. Go to Bard and open a new chat. Ask Bard to give you a list of PDFs in your drive. Don't ask it to access a folder because apparently it doesn't recognize folders yet. Bard will give you a list of PDFs in your drive. Ask it to summarize one of the articles in the list.

3. Go to Bard and open a new chat. Ask Bard to give you a list of PDFs in your drive. Don't ask it to access a folder because apparently it doesn't recognize folders yet. Bard will give you a list of PDFs in your drive. Ask it to summarize one of the articles in the list.

Mushtaq Bilal, PhD

25,012 次观看 • 2 年前

It may feel like you’re lacking; it’s over your head. That’s a sign that a destiny gene is about to open up. God will never ask you to do something and then not give you the ability to do it. Check out Joel's message, "Programmed for Greatness":

It may feel like you’re lacking; it’s over your head. That’s a sign that a destiny gene is about to open up. God will never ask you to do something and then not give you the ability to do it. Check out Joel's message, "Programmed for Greatness":

Joel Osteen

38,649 次观看 • 1 年前

He constantly was asking to be hit, so he got it! Be careful what you ask for. You may get it.

He constantly was asking to be hit, so he got it! Be careful what you ask for. You may get it.

KO Sequence

20,112 次观看 • 2 个月前

your coding agent sucks because it lacks context. so I built nia to fix that. here is how you can 10x your coding agent (thread):

your coding agent sucks because it lacks context. so I built nia to fix that. here is how you can 10x your coding agent (thread):

Arlan

54,135 次观看 • 11 个月前

Kenyon Martin on Kevin Durant comments about him saying he’s got to get out the way and get in the corner: “I’m going to do you a favor KD, I’m going to finish the sentence for you. You need to go ask Ime Udoka. They need to go ask the coach, I’ll finish it since you can’t say it, I’ll finish it for you. Go ask Ime why they got me in this situation constantly throughout the season. Go ask him why we don’t have no plays.” (Via Gil’s Arena)

Kenyon Martin on Kevin Durant comments about him saying he’s got to get out the way and get in the corner: “I’m going to do you a favor KD, I’m going to finish the sentence for you. You need to go ask Ime Udoka. They need to go ask the coach, I’ll finish it since you can’t say it, I’ll finish it for you. Go ask Ime why they got me in this situation constantly throughout the season. Go ask him why we don’t have no plays.” (Via Gil’s Arena)

NBA Courtside

692,838 次观看 • 3 个月前

Introducing Vet: catch your coding agent’s mistakes Open source, fast, local. Zero telemetry Can review full PRs and it tells us when Claude/Codex/OpenCode lies about passing tests it never ran 👿 We use it daily and it saves tons of headache

Introducing Vet: catch your coding agent’s mistakes Open source, fast, local. Zero telemetry Can review full PRs and it tells us when Claude/Codex/OpenCode lies about passing tests it never ran 👿 We use it daily and it saves tons of headache

Imbue

586,132 次观看 • 4 个月前

Let me save you hours of testing frontends. If you're ever working on a front-end, instead of writing tests, and adding puppeteer slop to your repo 1. Get an llm to write you with whatever needs to be tested 2. Copy that, go to browser 3. Open localhost with your selected app 4. Use Claude Chrome Extension or Parchi 5. Send it the prompt 6. QA engineering, there you go. Use models results and pass it back to your coding agent to fix whatever is flagged.

Let me save you hours of testing frontends. If you're ever working on a front-end, instead of writing tests, and adding puppeteer slop to your repo 1. Get an llm to write you with whatever needs to be tested 2. Copy that, go to browser 3. Open localhost with your selected app 4. Use Claude Chrome Extension or Parchi 5. Send it the prompt 6. QA engineering, there you go. Use models results and pass it back to your coding agent to fix whatever is flagged.

0xSero

109,943 次观看 • 5 个月前

🔎🤖LangSmith Insights Agent Really excited to launch our first in-product agent This agent lives inside LangSmith and combs through traces, giving you insights into: 🧑‍🤝‍🧑how users are using your agent ⁉️how your agent may be messing up 🛃{your custom insight here} The problem we saw was that people were launching agents... and didn't know how their users were actually using them! You put a chat box in front of people, and they may ask it anything - the surface area for agents is often super wide In addition - agents would fail silently. They could give a bad response - this wouldn't show up in error logs, but its good to know. If you know what look for, you can set up LLM as a judge evaluators. But what if you don't? (most people don't initially) The best way to figure this out - as Hamel Husain says - "look at your data". But LLMs are really good at looking at your data! So can they do it for you? This is exactly what insights agent attempts to do. It's live in LangSmith today. You can read more about it here:

🔎🤖LangSmith Insights Agent Really excited to launch our first in-product agent This agent lives inside LangSmith and combs through traces, giving you insights into: 🧑‍🤝‍🧑how users are using your agent ⁉️how your agent may be messing up 🛃{your custom insight here} The problem we saw was that people were launching agents... and didn't know how their users were actually using them! You put a chat box in front of people, and they may ask it anything - the surface area for agents is often super wide In addition - agents would fail silently. They could give a bad response - this wouldn't show up in error logs, but its good to know. If you know what look for, you can set up LLM as a judge evaluators. But what if you don't? (most people don't initially) The best way to figure this out - as Hamel Husain says - "look at your data". But LLMs are really good at looking at your data! So can they do it for you? This is exactly what insights agent attempts to do. It's live in LangSmith today. You can read more about it here:

Harrison Chase

98,520 次观看 • 8 个月前

In Cursor you can *easily* turn comments into code. Just write in natural language what you want in your file. Then highlight it, use CMD+K, and ask the AI to translate it into code - and it will! Turn your thoughts right into apps.

In Cursor you can easily turn comments into code. Just write in natural language what you want in your file. Then highlight it, use CMD+K, and ask the AI to translate it into code - and it will! Turn your thoughts right into apps.

Mckay Wrigley

106,485 次观看 • 1 年前