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。

Letting an LLM run code inside your app sounds reckless. Legion makes it boring. Lua VM in pure The Elixir programming language, timeout and memory budgets, and the only way out is the modules you hand it. Watch one rebuild its own landing page – link below 👇

Elixir by Software Mansion

2,542 subscribers

12,789 次观看 • 2 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

8 条评论

Elixir by Software Mansion 的头像

Elixir by Software Mansion2 个月前

Try it out:

Dima Mikielewicz 的头像

Dima Mikielewicz2 个月前

@elixirlang And a massive shoutout to @tvlabs and @davydog187 for building an amazing Lua VM in Elixir, without which the sandbox would never have been this secure!

Nicholas Price 的头像

Nicholas Price1 个月前

@elixirlang This is sooo wild holy fuck. Put this on product hunt and hacker news please.

Josef Richter 的头像

Josef Richter1 个月前

@elixirlang Nice! I did almost the same thing the other day, with a WASM sandbox instead of Lua:

Nicholas Price 的头像

Nicholas Price1 个月前

@elixirlang @ZachSDaniel1 you see this?!?

FaizanN 的头像

FaizanN1 个月前

@elixirlang Apart from lua, is it more secure than Dune which runs elixir in the sandbox?

aziz abdullaev 的头像

aziz abdullaev1 个月前

@elixirlang Sort of a sandbox inside a browser? Insane

Dimitri Semenovich 的头像

Dimitri Semenovich1 个月前

@elixirlang Why Lua? Please build this for Erlang and ultimately Elixir itself

相关视频

This guy explained agent harness, memory, and loop engineering in one video - clearer than most paid courses a raw LLM with no harness is a wild horse. all power, no direction > trace the run -> score it with an LLM -> diagnose the bottleneck -> patch it -> ship it that loop is what lets an agent get sharper on its own, run after run memory + harness + loop engineering + evals - that's the whole stack in one picture watch it, then grab the diagram below

This guy explained agent harness, memory, and loop engineering in one video - clearer than most paid courses a raw LLM with no harness is a wild horse. all power, no direction > trace the run -> score it with an LLM -> diagnose the bottleneck -> patch it -> ship it that loop is what lets an agent get sharper on its own, run after run memory + harness + loop engineering + evals - that's the whole stack in one picture watch it, then grab the diagram below

Mr. Buzzoni

45,278 次观看 • 3 个月前

Next. Notice the landing page on the letter Click on it, turns out, it’s not a landing page at all. It directs to an article. How is it one of the world’s top 3 communications schools makes such a huge error about its own landing page?

Next. Notice the landing page on the letter Click on it, turns out, it’s not a landing page at all. It directs to an article. How is it one of the world’s top 3 communications schools makes such a huge error about its own landing page?

Abhijit Iyer-Mitra

43,614 次观看 • 2 个月前

For those who missed the news: we made Elixir run in browser – and named it Popcorn🍿 With Popcorn, you can run and compile Elixir code in-browser, play with IEx, and even use OTP apps. And yes – it’s pretty fast. Sounds fun? Play with a live demo – link below 👇

For those who missed the news: we made Elixir run in browser – and named it Popcorn🍿 With Popcorn, you can run and compile Elixir code in-browser, play with IEx, and even use OTP apps. And yes – it’s pretty fast. Sounds fun? Play with a live demo – link below 👇

Elixir by Software Mansion

14,561 次观看 • 1 年前

I made Apple's App Store is obsolete. I built an Claude Code for your iPhone, so you can build and run whatever app you want all on your device! Just describe an app and it writes the Swift, runs it, reads its own errors, and fixes them. Then it lands on your app's home screen. Here's exactly how it works 👇

I made Apple's App Store is obsolete. I built an Claude Code for your iPhone, so you can build and run whatever app you want all on your device! Just describe an app and it writes the Swift, runs it, reads its own errors, and fixes them. Then it lands on your app's home screen. Here's exactly how it works 👇

JC Builds

34,514 次观看 • 4 个月前

Jev founder, Diogo Almeida, just put out a PDF on building a Jev harness for coding agents the claim: 200x faster, 400x cheaper the model stopped being the bottleneck a while ago. the harness around it is where the cost and the speed actually live five decisions sit on every loop, and only one of them writes code hand this PDF and the article below to your Claude Code or Codex instance and let it rebuild its own setup your agent gets a better harness tonight, and monday looks different than it would have 👇

Jev founder, Diogo Almeida, just put out a PDF on building a Jev harness for coding agents the claim: 200x faster, 400x cheaper the model stopped being the bottleneck a while ago. the harness around it is where the cost and the speed actually live five decisions sit on every loop, and only one of them writes code hand this PDF and the article below to your Claude Code or Codex instance and let it rebuild its own setup your agent gets a better harness tonight, and monday looks different than it would have 👇

Hanako

37,887 次观看 • 1 个月前

Ever wanted your AI Agent to create its own tools on-the-fly? Well, you can do it quickly, safely and securely with Durable Object Facets! The app starts as nothing more than an interface to an LLM. The LLM is able to write its own tools to add to its capabilities per request

Ever wanted your AI Agent to create its own tools on-the-fly? Well, you can do it quickly, safely and securely with Durable Object Facets! The app starts as nothing more than an interface to an LLM. The LLM is able to write its own tools to add to its capabilities per request

Ashley Peacock

33,382 次观看 • 6 个月前

Why Matteo might just win with Muse “As software, it is very, very good. It instantly works. This is the definition of great software. Meta already has the infrastructure, and this is running on its own LLM. From an infrastructure perspective, almost no one can compete. That is why it is fast. That is why it works well. You get all the VM, all the infrastructure, all the storage, and they have their own LLM.” Jason ✨👾SaaStr.Ai✨ Lemkin Love to hear your thoughts Tarek Sheasha Ryan Fox Harrison Chase Erik Bernhardsson

Why Matteo might just win with Muse “As software, it is very, very good. It instantly works. This is the definition of great software. Meta already has the infrastructure, and this is running on its own LLM. From an infrastructure perspective, almost no one can compete. That is why it is fast. That is why it works well. You get all the VM, all the infrastructure, all the storage, and they have their own LLM.” Jason ✨👾SaaStr.Ai✨ Lemkin Love to hear your thoughts Tarek Sheasha Ryan Fox Harrison Chase Erik Bernhardsson

Harry Stebbings

24,252 次观看 • 1 个月前

Today we're launching the new Wonder! The fastest way to go from design to code and back. Connect your code. Run your app locally. Turn screens into editable layers. Ship directly from the canvas. You can finally use your own Claude and Codex subscription. Try it out today.

Today we're launching the new Wonder! The fastest way to go from design to code and back. Connect your code. Run your app locally. Turn screens into editable layers. Ship directly from the canvas. You can finally use your own Claude and Codex subscription. Try it out today.

Wonder

311,678 次观看 • 6 天前

STOP COPY-PASTING BETWEEN YOUR AIs. NOW THEY TALK TO EACH OTHER. Until today you were the messenger. Copy from Grok, paste into Claude, copy from Claude, repeat. Bolter just fixed that. Here's how they connect: 1. Copy one line 2. Paste it into your AI 3. Confirm with a code 4. It's in the chat Then you share its handle and it starts working with your other AIs and your friends' AIs. Give them a task and they split it up. One picks the landing site, another designs the wheels, another writes the code. The rover is ready and you didn't copy a thing. Each AI keeps its memory and you approve every connection. Link below 👇🏼

STOP COPY-PASTING BETWEEN YOUR AIs. NOW THEY TALK TO EACH OTHER. Until today you were the messenger. Copy from Grok, paste into Claude, copy from Claude, repeat. Bolter just fixed that. Here's how they connect: 1. Copy one line 2. Paste it into your AI 3. Confirm with a code 4. It's in the chat Then you share its handle and it starts working with your other AIs and your friends' AIs. Give them a task and they split it up. One picks the landing site, another designs the wheels, another writes the code. The rover is ready and you didn't copy a thing. Each AI keeps its memory and you approve every connection. Link below 👇🏼

marcus

70,065 次观看 • 20 天前

The easiest way to find out which models you can run on your computer: Magnitude is a 100% free, open-source desktop app that profiles your machine and ranks the models across: - Speed - Accuracy - Intelligence - Memory required Then it downloads the one you pick and tunes it for your machine. Finally, connect your favorite harness (Pi, OpenCode, Claude Code, Codex, etc.) in one click. Get started here: (don't forget to star 🌟) My co-founder wrote a detailed article on running your favorite harnesses with local models. The article is quoted below.

The easiest way to find out which models you can run on your computer: Magnitude is a 100% free, open-source desktop app that profiles your machine and ranks the models across: - Speed - Accuracy - Intelligence - Memory required Then it downloads the one you pick and tunes it for your machine. Finally, connect your favorite harness (Pi, OpenCode, Claude Code, Codex, etc.) in one click. Get started here: (don't forget to star 🌟) My co-founder wrote a detailed article on running your favorite harnesses with local models. The article is quoted below.

Avi Chawla

15,465 次观看 • 1 个月前

Remind that the best drawing software out rn is 50% percent off! And unlike other drawing programs when you buy CSP you actually own it! If your interested giving it a test run (and helping me out) I have an affiliate link below🔽

Remind that the best drawing software out rn is 50% percent off! And unlike other drawing programs when you buy CSP you actually own it! If your interested giving it a test run (and helping me out) I have an affiliate link below🔽

Seth Redd WISHLIST SKINS GAME/Commissions Open

259,178 次观看 • 7 个月前

This idea is either extremely smart or an extremely stupid—no in-between. What if your LLM *is* your search engine? How would you look like inside it? Forget about Perplexity, DeepResearch. What if LLM is your entire Google? Pagination, links and everything - just like the old days: no chat UI, classic Google vibe. If you're unsure what's that mean, watch the demo video below first. We call it LLM-as-SERP (Search Engine Results Page).

This idea is either extremely smart or an extremely stupid—no in-between. What if your LLM is your search engine? How would you look like inside it? Forget about Perplexity, DeepResearch. What if LLM is your entire Google? Pagination, links and everything - just like the old days: no chat UI, classic Google vibe. If you're unsure what's that mean, watch the demo video below first. We call it LLM-as-SERP (Search Engine Results Page).

Jina AI

73,960 次观看 • 1 年前

It is becoming f*cking dangerous: Claude Code has kept its own memory folder since February and you've probably never opened it 1. Open any project in Claude Code and type /memory 2. Check the auto memory toggle, it is on by default in local sessions 3. Pick the option to open the auto memory folder 4. Read MEMORY.md first, it is the index, one line per memory 5. Look at the files: one memory each, typed user, feedback, project or reference 6. Tell Claude "remember: always use pnpm, not npm" and watch it save a new memory 7. Edit or delete any wrong memory, it is plain markdown Only the first 200 lines or 25KB of MEMORY.md load each session, Claude opens the topic files when needed Ask Claude to remember what matters once, it stays until you or Claude edit or delete it Full guide in the article below

It is becoming f*cking dangerous: Claude Code has kept its own memory folder since February and you've probably never opened it 1. Open any project in Claude Code and type /memory 2. Check the auto memory toggle, it is on by default in local sessions 3. Pick the option to open the auto memory folder 4. Read MEMORY.md first, it is the index, one line per memory 5. Look at the files: one memory each, typed user, feedback, project or reference 6. Tell Claude "remember: always use pnpm, not npm" and watch it save a new memory 7. Edit or delete any wrong memory, it is plain markdown Only the first 200 lines or 25KB of MEMORY.md load each session, Claude opens the topic files when needed Ask Claude to remember what matters once, it stays until you or Claude edit or delete it Full guide in the article below

plusxbt

19,250 次观看 • 4 天前

Our NFT Heroes have been revealed. The Mobile App is set to launch. Today is our first art reveal for The Code. Watch the video below and screenshot your fav! Share it in replies 👇

Our NFT Heroes have been revealed. The Mobile App is set to launch. Today is our first art reveal for The Code. Watch the video below and screenshot your fav! Share it in replies 👇

GameGPT

57,458 次观看 • 1 年前

Me and You Out Now world wide let’s run it up,tap on the link below🔥👇

Me and You Out Now world wide let’s run it up,tap on the link below🔥👇

Triple 4K

426,736 次观看 • 3 年前

You can now run MicroPython on an emulated Raspberry Pi Pico - inside a real circuit simulator, on your iPhone, iPad, and Mac. Write the code, wire the GPIOs into analog + digital circuits, and watch it all run. Live. iCircuit 1.16 beta is ready. Testers wanted 👇

You can now run MicroPython on an emulated Raspberry Pi Pico - inside a real circuit simulator, on your iPhone, iPad, and Mac. Write the code, wire the GPIOs into analog + digital circuits, and watch it all run. Live. iCircuit 1.16 beta is ready. Testers wanted 👇

Frank A. Krueger

11,129 次观看 • 3 个月前

Wow... This Claude Code Design Hack is wild. When building landing pages, don't ask for a single landing page. Ask for a directory of landing page options... and in your 1st prompt tell it to learn from your preferences as you generate more. Use this prompt below 👇

Wow... This Claude Code Design Hack is wild. When building landing pages, don't ask for a single landing page. Ask for a directory of landing page options... and in your 1st prompt tell it to learn from your preferences as you generate more. Use this prompt below 👇

Learn to Vibecode

187,098 次观看 • 10 个月前

Maybe I only find this interesting but the Jags call timeout just before the snap and CJ Stroud still throws it to Foster Moreau and he casually catches the ball with one hand. And then they run the real play after the timeout...

Maybe I only find this interesting but the Jags call timeout just before the snap and CJ Stroud still throws it to Foster Moreau and he casually catches the ball with one hand. And then they run the real play after the timeout...

Goodberry

510,132 次观看 • 15 天前

your Hermes Agent, on your wrist. Hermes Watch is a new open-source community build for the Amazfit Active 2 watch: tap TALK, ask, and Hermes Agent answers on the watch, in text and out loud. what it does: - voice in, answers on your wrist, no phone in your hand - reads answers out loud, with a Voice on/off button - one ongoing conversation until you say "new chat" - Approve / Deny cards when Hermes asks before running something, one request at a time - long jobs keep running after you close the app, and the result lands on your Telegram - texts, car commands and a PC restart only run after you tap Yes - runs on your own PC, reachable only inside your Tailscale network what you can do with it: - ask a quick question while you walk or cook - approve the code change your agent wants to make, away from your desk - start a build and get pinged when it's done - check on your PC with a quick health check opensource code below

your Hermes Agent, on your wrist. Hermes Watch is a new open-source community build for the Amazfit Active 2 watch: tap TALK, ask, and Hermes Agent answers on the watch, in text and out loud. what it does: - voice in, answers on your wrist, no phone in your hand - reads answers out loud, with a Voice on/off button - one ongoing conversation until you say "new chat" - Approve / Deny cards when Hermes asks before running something, one request at a time - long jobs keep running after you close the app, and the result lands on your Telegram - texts, car commands and a PC restart only run after you tap Yes - runs on your own PC, reachable only inside your Tailscale network what you can do with it: - ask a quick question while you walk or cook - approve the code change your agent wants to make, away from your desk - start a build and get pinged when it's done - check on your PC with a quick health check opensource code below

witcheer

15,273 次观看 • 10 天前