The best Hacker News stories from Show from the past day
Latest posts:
Show HN: Hacker News frontpage as a print newspaper that you can personalize
Show HN: Hacker News frontpage as a print newspaper that you can personalize
Show HN: Hacker News frontpage as a print newspaper that you can personalize
Show HN: Hacker News frontpage as a print newspaper that you can personalize
Rd-TableBench – Accurately evaluating table extraction
Hey HN!<p>A ton of document parsing solutions have been coming out lately, each claiming SOTA with little evidence. A lot of these turned out to be LLM or LVM wrappers that hallucinate frequently on complex tables.<p>We just released RD-TableBench, an open benchmark to help teams evaluate extraction performance for complex tables. The benchmark includes a variety of challenging scenarios including scanned tables, handwriting, language detection, merged cells, and more.<p>We employed an independent team of PhD-level human labelers who manually annotated 1000 complex table images from a diverse set of publicly available documents.<p>Alongside this, we also release a new bioinformatics inspired algorithm for grading table similarity. Would love to hear any feedback!<p>-Raunak
Show HN: Varse – Simple remote application config
Hey HN,<p>We wanted a simple way to update application configs, without redeploying. We wanted to remotely set variables, and read them in our application. We created Varse to do this.<p>Varse has a dashboard for creating key - value pairs and an SDK to read them. It's un-opinionated and allows for strings, booleans, or even json objects to be stored.<p>We have instructions for running it yourself. We also have a hosted version where you can create an account and manage variables.<p>This is our first time building an open source project. We'd love feedback on how to do it right.<p>Github: <a href="https://github.com/varse-io/varse">https://github.com/varse-io/varse</a><p>Hosted Version: <a href="https://app.varse.io/signup" rel="nofollow">https://app.varse.io/signup</a><p>Website: <a href="https://www.varse.io/" rel="nofollow">https://www.varse.io/</a><p>Contact: izak@varse.io
Show HN: Varse – Simple remote application config
Hey HN,<p>We wanted a simple way to update application configs, without redeploying. We wanted to remotely set variables, and read them in our application. We created Varse to do this.<p>Varse has a dashboard for creating key - value pairs and an SDK to read them. It's un-opinionated and allows for strings, booleans, or even json objects to be stored.<p>We have instructions for running it yourself. We also have a hosted version where you can create an account and manage variables.<p>This is our first time building an open source project. We'd love feedback on how to do it right.<p>Github: <a href="https://github.com/varse-io/varse">https://github.com/varse-io/varse</a><p>Hosted Version: <a href="https://app.varse.io/signup" rel="nofollow">https://app.varse.io/signup</a><p>Website: <a href="https://www.varse.io/" rel="nofollow">https://www.varse.io/</a><p>Contact: izak@varse.io
Show HN: rallyup – Lightweight Wake-on-LAN Scheduler
Hi HN,<p>I’ve wanted a simple solution to handle Wake-on-LAN sequences for my home and work labs to boot up servers in the right order. I was already dabbling in Rust and thought this would be an interesting project to dive deeper and see if it could work well for this kind of network tool. The result is rallyup.<p>rallyup lets you set up server dependencies in a YAML file, so each service (e.g., firewalls, storage, VM hosts) comes online in the right order. It verifies each server’s status before moving to the next.<p>Features:<p>- Dependency-based WOL with VLAN support
- Built-in health checks (HTTP, open ports, shell commands)
- Lightweight enough to run on a Raspberry Pi or similar device<p>Would love any feedback. Thanks for taking a look!
Show HN: rallyup – Lightweight Wake-on-LAN Scheduler
Hi HN,<p>I’ve wanted a simple solution to handle Wake-on-LAN sequences for my home and work labs to boot up servers in the right order. I was already dabbling in Rust and thought this would be an interesting project to dive deeper and see if it could work well for this kind of network tool. The result is rallyup.<p>rallyup lets you set up server dependencies in a YAML file, so each service (e.g., firewalls, storage, VM hosts) comes online in the right order. It verifies each server’s status before moving to the next.<p>Features:<p>- Dependency-based WOL with VLAN support
- Built-in health checks (HTTP, open ports, shell commands)
- Lightweight enough to run on a Raspberry Pi or similar device<p>Would love any feedback. Thanks for taking a look!
Show HN: Whirlwind – Async concurrent hashmap for Rust
Hey HN, this is Will and David from Fortress (<a href="https://news.ycombinator.com/item?id=41426998">https://news.ycombinator.com/item?id=41426998</a>).<p>We use a lot of async Rust internally, and created this library out of a need for an async-aware concurrent hashmap since there weren’t many available in the Rust ecosystem.<p>Whirlwind is a sharded HashMap with a fully asynchronous API. Just as dashmap is a replacement for std::sync::RwLock<HashMap>, whirlwind aims to be a replacement for tokio::sync::RwLock<HashMap>. It has a similar design and performance characteristics to dashmap, but seems to perform better in read-heavy workloads with tokio's green threading.<p>Benchmarks are in the readme! We used an asyncified version of dashmap's benchmark suite. The project is in a pretty early stage and I'm sure there are flaws, but I'm pretty happy with the performance.<p>There is some unsafe involved, but we run Miri in ci to (hopefully) catch undefined behavior well before it's in an actual release.<p>We'd appreciate any feedback! Thanks in advance :)
Show HN: Whirlwind – Async concurrent hashmap for Rust
Hey HN, this is Will and David from Fortress (<a href="https://news.ycombinator.com/item?id=41426998">https://news.ycombinator.com/item?id=41426998</a>).<p>We use a lot of async Rust internally, and created this library out of a need for an async-aware concurrent hashmap since there weren’t many available in the Rust ecosystem.<p>Whirlwind is a sharded HashMap with a fully asynchronous API. Just as dashmap is a replacement for std::sync::RwLock<HashMap>, whirlwind aims to be a replacement for tokio::sync::RwLock<HashMap>. It has a similar design and performance characteristics to dashmap, but seems to perform better in read-heavy workloads with tokio's green threading.<p>Benchmarks are in the readme! We used an asyncified version of dashmap's benchmark suite. The project is in a pretty early stage and I'm sure there are flaws, but I'm pretty happy with the performance.<p>There is some unsafe involved, but we run Miri in ci to (hopefully) catch undefined behavior well before it's in an actual release.<p>We'd appreciate any feedback! Thanks in advance :)
Show HN: Whirlwind – Async concurrent hashmap for Rust
Hey HN, this is Will and David from Fortress (<a href="https://news.ycombinator.com/item?id=41426998">https://news.ycombinator.com/item?id=41426998</a>).<p>We use a lot of async Rust internally, and created this library out of a need for an async-aware concurrent hashmap since there weren’t many available in the Rust ecosystem.<p>Whirlwind is a sharded HashMap with a fully asynchronous API. Just as dashmap is a replacement for std::sync::RwLock<HashMap>, whirlwind aims to be a replacement for tokio::sync::RwLock<HashMap>. It has a similar design and performance characteristics to dashmap, but seems to perform better in read-heavy workloads with tokio's green threading.<p>Benchmarks are in the readme! We used an asyncified version of dashmap's benchmark suite. The project is in a pretty early stage and I'm sure there are flaws, but I'm pretty happy with the performance.<p>There is some unsafe involved, but we run Miri in ci to (hopefully) catch undefined behavior well before it's in an actual release.<p>We'd appreciate any feedback! Thanks in advance :)
Show HN: IMDb SQL Best Movie Finder
I've built a static web app called IMDb SQL Best Movie Finder that lets you query a database of 1.5 million IMDb titles using SQL directly in your browser. It’s entirely client-side, so all the data processing happens locally on your machine — no server involved.
Show HN: I wrote an open-source browser alternative for Computer Use for any LLM
Hey HN,<p>I made Browser-Use, an open-source tool that lets (all Langchain supported) LLMs execute tasks directly in the browser just with function calling.<p>It allows you to build agents that interact with web elements using natural language prompts. We created a layer that simplifies website interaction for LLMs by extracting xPaths and interactive elements like buttons and input fields (and other fancy things). This enables you to design custom web automation and scraping functions without manual inspection through DevTools.<p>Hasn't this been done a lot of times?
Good question, as a general SaaS tool yes, but I think a lot of people are going to try to make their own web automation agents from scratch, so the idea is to provide groundwork/library for the hard part so that not everyone has to repeat these steps:<p>- parse html in a LLM friendly way (clickable items + screenshots)<p>- provide a nice function calls for everything inside the browser<p>- create reusable agent classes<p>What this is NOT? An all knowing AI agent that can solve all your problems.<p>The vision: create repeatable tasks on the web just by prompting your agent and not care about the hows.<p>To better showcase the power of text extraction we made a few demos such as:<p>- Applying for multiple software engineering jobs in San Francisco<p>- Opening new tabs to search for images of Albert Einstein, Oprah Winfrey, and Steve Jobs<p>- Finding the cheapest one-way flight from London to Kyrgyzstan for December 25th<p>I’d be interested in feedback on how this tool fits into your automation workflows. Try it out and let me know how it performs on your end.<p>We are Gregor & Magnus and we built this in 5 days.
Show HN: I wrote an open-source browser alternative for Computer Use for any LLM
Hey HN,<p>I made Browser-Use, an open-source tool that lets (all Langchain supported) LLMs execute tasks directly in the browser just with function calling.<p>It allows you to build agents that interact with web elements using natural language prompts. We created a layer that simplifies website interaction for LLMs by extracting xPaths and interactive elements like buttons and input fields (and other fancy things). This enables you to design custom web automation and scraping functions without manual inspection through DevTools.<p>Hasn't this been done a lot of times?
Good question, as a general SaaS tool yes, but I think a lot of people are going to try to make their own web automation agents from scratch, so the idea is to provide groundwork/library for the hard part so that not everyone has to repeat these steps:<p>- parse html in a LLM friendly way (clickable items + screenshots)<p>- provide a nice function calls for everything inside the browser<p>- create reusable agent classes<p>What this is NOT? An all knowing AI agent that can solve all your problems.<p>The vision: create repeatable tasks on the web just by prompting your agent and not care about the hows.<p>To better showcase the power of text extraction we made a few demos such as:<p>- Applying for multiple software engineering jobs in San Francisco<p>- Opening new tabs to search for images of Albert Einstein, Oprah Winfrey, and Steve Jobs<p>- Finding the cheapest one-way flight from London to Kyrgyzstan for December 25th<p>I’d be interested in feedback on how this tool fits into your automation workflows. Try it out and let me know how it performs on your end.<p>We are Gregor & Magnus and we built this in 5 days.
Show HN: Mem0 Browser Extension: Shared Memory Across ChatGPT,Claude,Perplexity
Hey HN! We're Deshraj and Taranjeet. We've been building working on a startup called Mem0, building an open-source memory layer for AI apps and agents (<a href="https://news.ycombinator.com/item?id=41447317">https://news.ycombinator.com/item?id=41447317</a>). We also kept running into our own daily frustrations with AI assistants forgetting everything between conversations. Over a weekend, we decided to hack together a Chrome extension to solve this for ourselves.<p>The problem was simple: we were constantly re-explaining our context across platforms when switching between ChatGPT, Claude, and Perplexity. Start a coding discussion in ChatGPT, switch to Claude for a different perspective, jump to Perplexity for research—you're starting from scratch each time. We thought others might find this useful too, so we're sharing it with the HN community.<p>Our solution is built on our unified memory layer that works across multiple LLMs, accessible through a simple Chrome extension. Here’s a quick demo of how it works: <a href="https://youtu.be/cByzXztn-YY" rel="nofollow">https://youtu.be/cByzXztn-YY</a><p>The key features include<p>- Cross-LLM Memory: Start a conversation in ChatGPT, then continue in Claude or Perplexity without losing your context. This makes it easy to switch between models while maintaining coherence in your interactions.<p>- Customizable Control: Our dashboard lets you manage memories directly—you can add, edit, or delete memories, ensuring that your context stays relevant and accurate across all your LLM interactions.<p>- Sync with ChatGPT Memories: If you've been using ChatGPT's memory feature, Mem0 can sync with it, creating a consistent experience across your preferred LLMs.<p>We use a hybrid data architecture that combines graph, vector, and key-value stores to manage memories. This setup enables efficient memory retrieval based on relevance, recency, and context, ensuring your interactions remain meaningful across all apps.<p>The Chrome extension is MIT licensed and available on GitHub. Currently, it uses our hosted version of Mem0 for simplicity and stability. But we plan to add support for self-hosting using the open-source version of Mem0.<p>Try it out:<p>- Extension: <a href="https://mem0.dev/extension" rel="nofollow">https://mem0.dev/extension</a>
- Source code: <a href="https://github.com/mem0ai/mem0-chrome-extension">https://github.com/mem0ai/mem0-chrome-extension</a><p>We'd love to hear any feedback and suggestions!
Show HN: SplatGallery – A Community-Driven Gallery for Gaussian Splats
I just launched SplatGallery, a platform where you can share and discover the best 3D models created with Gaussian Splatting.<p>It's a super early stage and new models are coming it fairly often.<p>Would love to get feedback from the HN community on how to improve it!
Show HN: Convert any website into a React component
Hey HN, we built a Chrome Extension (<a href="https://www.magicpatterns.com/extension">https://www.magicpatterns.com/extension</a>) that converts a snippet of any website to an isolated React component.<p>Demo video: <a href="https://youtu.be/hutUYDkyE_A" rel="nofollow">https://youtu.be/hutUYDkyE_A</a><p>How it works: 1) Iterate through each node in the selected DOM tree, 2) For each element, find any matching CSS selectors / inline styles, 3) Use window.getComputedStyle to get the deterministic values, 4) Construct JSX<p>It was pretty hard producing the minimal code necessary while maintaining the same visual look. To do this, we implemented things like abstracting out global styles, removing inherited styles, pulling out SVGs, deleting styles with no effect, and condensing styles into their shorthand properties.<p>We dive into each of those optimizations here for fun: <a href="https://www.magicpatterns.com/blog/any-website-to-react-component">https://www.magicpatterns.com/blog/any-website-to-react-comp...</a><p>One of the main reasons we cared so much about condensing down the styles was not only to make it more human-readable, but also to reduce context length for an LLM, so that you can iterate on it with AI. Our extension has a “convert” option that lets you convert the output to Tailwind, Shadcn, or Chakra UI using an LLM. You can also export to Figma.<p>We're frontend engineers and we built the extension because our core product (<a href="https://www.magicpatterns.com/">https://www.magicpatterns.com/</a>) helps software teams prototype their product ideas. And a huge pain point for users is getting their existing designs into our product, so that they can reference them to generate UIs with their existing aesthetic.<p>The extension allows you to get existing design context from any website, even localhost. Since launch, the extension has more than 3,000 users and interestingly is most popular in Japan.<p>Here's some real examples if you're curious what the final output looks like:<p>A) Hacker News Navbar - <a href="https://www.magicpatterns.com/ac9f38e4-5ef0-49e5-8b80-dbc42951a00a">https://www.magicpatterns.com/ac9f38e4-5ef0-49e5-8b80-dbc429...</a><p>B) ChatGPT Welcome Screen - <a href="https://www.magicpatterns.com/7cb3ad12-cb12-4a5b-b32b-eda04de9ec01">https://www.magicpatterns.com/7cb3ad12-cb12-4a5b-b32b-eda04d...</a><p>C) Cal.com Calendar Component — <a href="https://www.magicpatterns.com/a43bac78-134d-458d-8107-811ac7b32b1f">https://www.magicpatterns.com/a43bac78-134d-458d-8107-811ac7...</a><p>D) Stripe.com logo section - <a href="https://www.magicpatterns.com/deff1793-7a05-42fe-97f7-945976cdbc7e">https://www.magicpatterns.com/deff1793-7a05-42fe-97f7-945976...</a><p>If you have an opinion about the extension, we’re all ears! You can try it for free at: <a href="https://www.magicpatterns.com/extension">https://www.magicpatterns.com/extension</a>
Show HN: Convert any website into a React component
Hey HN, we built a Chrome Extension (<a href="https://www.magicpatterns.com/extension">https://www.magicpatterns.com/extension</a>) that converts a snippet of any website to an isolated React component.<p>Demo video: <a href="https://youtu.be/hutUYDkyE_A" rel="nofollow">https://youtu.be/hutUYDkyE_A</a><p>How it works: 1) Iterate through each node in the selected DOM tree, 2) For each element, find any matching CSS selectors / inline styles, 3) Use window.getComputedStyle to get the deterministic values, 4) Construct JSX<p>It was pretty hard producing the minimal code necessary while maintaining the same visual look. To do this, we implemented things like abstracting out global styles, removing inherited styles, pulling out SVGs, deleting styles with no effect, and condensing styles into their shorthand properties.<p>We dive into each of those optimizations here for fun: <a href="https://www.magicpatterns.com/blog/any-website-to-react-component">https://www.magicpatterns.com/blog/any-website-to-react-comp...</a><p>One of the main reasons we cared so much about condensing down the styles was not only to make it more human-readable, but also to reduce context length for an LLM, so that you can iterate on it with AI. Our extension has a “convert” option that lets you convert the output to Tailwind, Shadcn, or Chakra UI using an LLM. You can also export to Figma.<p>We're frontend engineers and we built the extension because our core product (<a href="https://www.magicpatterns.com/">https://www.magicpatterns.com/</a>) helps software teams prototype their product ideas. And a huge pain point for users is getting their existing designs into our product, so that they can reference them to generate UIs with their existing aesthetic.<p>The extension allows you to get existing design context from any website, even localhost. Since launch, the extension has more than 3,000 users and interestingly is most popular in Japan.<p>Here's some real examples if you're curious what the final output looks like:<p>A) Hacker News Navbar - <a href="https://www.magicpatterns.com/ac9f38e4-5ef0-49e5-8b80-dbc42951a00a">https://www.magicpatterns.com/ac9f38e4-5ef0-49e5-8b80-dbc429...</a><p>B) ChatGPT Welcome Screen - <a href="https://www.magicpatterns.com/7cb3ad12-cb12-4a5b-b32b-eda04de9ec01">https://www.magicpatterns.com/7cb3ad12-cb12-4a5b-b32b-eda04d...</a><p>C) Cal.com Calendar Component — <a href="https://www.magicpatterns.com/a43bac78-134d-458d-8107-811ac7b32b1f">https://www.magicpatterns.com/a43bac78-134d-458d-8107-811ac7...</a><p>D) Stripe.com logo section - <a href="https://www.magicpatterns.com/deff1793-7a05-42fe-97f7-945976cdbc7e">https://www.magicpatterns.com/deff1793-7a05-42fe-97f7-945976...</a><p>If you have an opinion about the extension, we’re all ears! You can try it for free at: <a href="https://www.magicpatterns.com/extension">https://www.magicpatterns.com/extension</a>
Show HN: Convert any website into a React component
Hey HN, we built a Chrome Extension (<a href="https://www.magicpatterns.com/extension">https://www.magicpatterns.com/extension</a>) that converts a snippet of any website to an isolated React component.<p>Demo video: <a href="https://youtu.be/hutUYDkyE_A" rel="nofollow">https://youtu.be/hutUYDkyE_A</a><p>How it works: 1) Iterate through each node in the selected DOM tree, 2) For each element, find any matching CSS selectors / inline styles, 3) Use window.getComputedStyle to get the deterministic values, 4) Construct JSX<p>It was pretty hard producing the minimal code necessary while maintaining the same visual look. To do this, we implemented things like abstracting out global styles, removing inherited styles, pulling out SVGs, deleting styles with no effect, and condensing styles into their shorthand properties.<p>We dive into each of those optimizations here for fun: <a href="https://www.magicpatterns.com/blog/any-website-to-react-component">https://www.magicpatterns.com/blog/any-website-to-react-comp...</a><p>One of the main reasons we cared so much about condensing down the styles was not only to make it more human-readable, but also to reduce context length for an LLM, so that you can iterate on it with AI. Our extension has a “convert” option that lets you convert the output to Tailwind, Shadcn, or Chakra UI using an LLM. You can also export to Figma.<p>We're frontend engineers and we built the extension because our core product (<a href="https://www.magicpatterns.com/">https://www.magicpatterns.com/</a>) helps software teams prototype their product ideas. And a huge pain point for users is getting their existing designs into our product, so that they can reference them to generate UIs with their existing aesthetic.<p>The extension allows you to get existing design context from any website, even localhost. Since launch, the extension has more than 3,000 users and interestingly is most popular in Japan.<p>Here's some real examples if you're curious what the final output looks like:<p>A) Hacker News Navbar - <a href="https://www.magicpatterns.com/ac9f38e4-5ef0-49e5-8b80-dbc42951a00a">https://www.magicpatterns.com/ac9f38e4-5ef0-49e5-8b80-dbc429...</a><p>B) ChatGPT Welcome Screen - <a href="https://www.magicpatterns.com/7cb3ad12-cb12-4a5b-b32b-eda04de9ec01">https://www.magicpatterns.com/7cb3ad12-cb12-4a5b-b32b-eda04d...</a><p>C) Cal.com Calendar Component — <a href="https://www.magicpatterns.com/a43bac78-134d-458d-8107-811ac7b32b1f">https://www.magicpatterns.com/a43bac78-134d-458d-8107-811ac7...</a><p>D) Stripe.com logo section - <a href="https://www.magicpatterns.com/deff1793-7a05-42fe-97f7-945976cdbc7e">https://www.magicpatterns.com/deff1793-7a05-42fe-97f7-945976...</a><p>If you have an opinion about the extension, we’re all ears! You can try it for free at: <a href="https://www.magicpatterns.com/extension">https://www.magicpatterns.com/extension</a>