The URL shortener that makes your links look as suspicious as possible

Raspberry Pi's New AI Hat Adds 8GB of RAM for Local LLMs

Claude Cowork exfiltrates files

Handy – free open source speech-to-text app

Furiosa: 3.5x efficiency over H100s

Have Taken Up Farming

For the past year I&#x27;ve been working to rethink how AI manages timing in conversation at Tavus. I&#x27;ve spent a lot of time listening to conversations. Today we&#x27;re announcing the release of Sparrow-1, the most advanced conversational flow model in the world.Some technical details:- Predicts conversational floor ownership, not speech endpoints- Audio-native streaming model, no ASR dependency- Human-timed responses without silence-based delays- Zero interruptions at sub-100ms median latency- In benchmarks Sparrow-1 beats all existing models at real world turn-taking baselinesI wrote more about the work here: <a href="https:&#x2F;&#x2F;www.tavus.io&#x2F;post&#x2F;sparrow-1-human-level-conversational-timing-in-real-time-voice">https:&#x2F;&#x2F;www.tavus.io&#x2F;post&#x2F;sparrow-1-human-level-conversation...</a>

Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

New Safari developer tools provide insight into CSS Grid Lanes

I am curious how people are doing RAG locally with minimal dependencies for internal code or complex documents?Are you using a vector database, some type of semantic search, a knowledge graph, a hypergraph?

Ask HN: How are you doing RAG locally?

The <Geolocation> HTML Element

Hello HN! I am putting together a community-maintained directory of personal websites at &lt;https:&#x2F;&#x2F;hnpwd.github.io&#x2F;&gt;. More details about the project can be found in the README at &lt;https:&#x2F;&#x2F;github.com&#x2F;hnpwd&#x2F;hnpwd.github.io#readme&gt;.As you can see, the directory currently has only a handful of entries. I need your help to grow it. If you have a personal website, I would be glad if you shared it here. If your website is hosted on a web space where you have full control over its design and content, and if it has been well received in past HN discussions, I might add it to the directory. Just drop a link in the comments. Please let me know if you do not want your website to be included in the directory.Also, I intend this to be a community maintained resource, so if you would like to join the GitHub project as a maintainer, please let me know either here or via the IRC link in the README.By the way, see also &#x27;Ask HN: Could you share your personal blog here?&#x27; - https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=36575081 - July 2023 - (1014 points, 1940 comments). In this post, the scope is not restricted to blogs though. Any personal website is welcome, whether it is a blog, digital garden, personal wiki or something else entirely.UPDATE: It is going to take a while to go through all the submissions and add them. If you&#x27;d like to help with the process, please send a PR directly to this project: https:&#x2F;&#x2F;github.com&#x2F;hnpwd&#x2F;hnpwd.github.io

Ask HN: Share your personal website

Scaling long-running autonomous coding

Bubblewrap: A nimble way to prevent agents from accessing your .env files

Bare metal programming with RISC-V guide (2023)

Doesn&#x27;t matter what domain and how big or small.

Ask HN: What did you find out or explore today?

The State of OpenSSL for pyca/cryptography

Project SkyWatch (a.k.a. Wescam at Home)

There is a large grid of 250x250 tiles, on which you are be able to create a tiny website, contained into the tile.
You can basically consider the tile as a mini version of your website, showcasing what your full site has (but it can be anything). You are able to link to your full site, and use any HTML&#x2F;CSS&#x2F;JS inside. The purpose is to create beautiful and interesting tiles, that could be used for exploring the indie-web in an easy and interesting way.

Show HN: WebTiles – create a tiny 250x250 website with neighbors around you

Sun Position Calculator

SparkFun Officially Dropping AdaFruit due to CoC Violation

Find a pub that needs you

Hi HN, I built webctl because I was frustrated by the gap between curl and full browser automation frameworks like Playwright.I initially built this to solve a personal headache: I wanted an AI agent to handle project management tasks on my company’s intranet. I needed it to persist cookies across sessions (to handle SSO) and then scrape a Kanban board.Existing AI browser tools (like current MCP implementations) often force unsolicited data into the context window—dumping the full accessibility tree, console logs, and network errors whether you asked for them or not.webctl is an attempt to solve this with a Unix-style CLI:- Filter before context: You pipe the output to standard tools. webctl snapshot --interactive-only | head -n 20 means the LLM only sees exactly what I want it to see.- Daemon Architecture: It runs a persistent background process. The goal is to keep the browser state (cookies&#x2F;session) alive while you run discrete, stateless CLI commands.- Semantic targeting: It uses ARIA roles (e.g., role=button name~=&quot;Submit&quot;) rather than fragile CSS selectors.Disclaimer: The daemon logic for state persistence is still a bit experimental, but the architecture feels like the right direction for building local, token-efficient agents.It’s basically &quot;Playwright for the terminal.&quot;

Show HN: Webctl – Browser automation for agents based on CLI instead of MCP

Crafting Interpreters

archive.today has recently (I noticed this, like, 3 days ago) started automatically making requests to someone&#x27;s personal blog on their CAPTCHA page. Here&#x27;s a screenshot of what I&#x27;m talking about: https:&#x2F;&#x2F;files.catbox.moe&#x2F;20jsle.pngThe relevant JS is:<pre><code> setInterval(function() {
 fetch(&quot;https:&#x2F;&#x2F;gyrovague.com&#x2F;?s=&quot; + Math.round(new Date().getTime() % 10000000), {
 referrerPolicy: &quot;no-referrer&quot;,
 mode: &quot;no-cors&quot;
 });
 }, 300);
</code></pre>
Looking at this blog, there seems to be exactly one article mentioning archive.today - &quot;archive.today: On the trail of the mysterious guerrilla archivist of the Internet&quot; (https:&#x2F;&#x2F;gyrovague.com&#x2F;2023&#x2F;08&#x2F;05&#x2F;archive-today-on-the-trail-of-the-mysterious-guerrilla-archivist-of-the-internet&#x2F;), where the person running the blog digs up some information about archive&#x27;s owner.So perhaps this is some kind of revenge&#x2F;DOS attack attempt&#x2F;deliberately wasting their bandwidth in response to this article? Maybe an attempt to silence them and force to delete their article? But if it is, then I have so many questions. Like, why would the owner of the archive do that 2.5 years after the article was published? Or why would they even do that in the first place, do they not know about Streisand effect?I&#x27;m confused.

Ask HN: Weird archive.today behavior?

Generate QR Codes with Pure SQL in PostgreSQL

How can I build a simple pulse generator to demonstrate transmission lines

MIT Whirlwind I: A High-Speed Electronic Digital Computer (1951) [pdf]

<a href="https:&#x2F;&#x2F;github.com&#x2F;stepandel&#x2F;chroma-explorer" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;stepandel&#x2F;chroma-explorer</a>

ChromaDB Explorer

With research done till date, what according to you is the best way to provide context to a model. Are there any articles that go into depth of how Cursor does it?How do context collation companies work?