23,447 results for sing · 2.476s

News for “sing”
20 results • 2471 ms server time
Moozonian News
news.ycombinator.com• Jul 22, 2026• 1 min read
New Inference Server for DGX Spark: large model C4:55-90 tok/s no spec decodeHi All, We are so excited to share the numbers and benchmark reports on our new inference server built specifically to run multi-model agentic workflows on DGX Spark clusters. We ran LlamaBench tests and also our own simulated traffic test on a 2 DGX Spark cluster setup and got some really good numbers. Here are the details. The full detailed report is available on our WoolyAI website.WoolyAI Private Multi-agent Inference Stack for DGX Spark is built to enable groups within enterprises to set up their own private, low-cost inference stacks for business agentic workflow apps. It was built with the forward-looking vision that companies will need their own private, low-cost inference setup and that single-model-based inference is not sufficient for complex enterprise workflow agentic apps. These workflows need multiple models of different specializations (hence sizes) for different steps in the workflows. Having dedicated multi-GPU inference stacks for each model is very cost-prohibitive.
Moozonian News
thedaily.fm• Jul 22, 2026• 1 min read
Show HN: The Daily FM – Turn any source into a daily podcastI built The Daily FM because I wanted a daily summary of the latest AI news sent to my podcast app and then realized it was useful in summarizing other stuff:- Any blogs, websites, or X handles- The top Hacker News stories with comments- Long podcasts (think Lex Fridman)- Changes in state/federal legislationIt pulls your sources, writes a script with a frontier-ish model of your choice, runs it through TTS (MAI-2 studio voices), and publishes a standard RSS feed. You can also combine several pods into one feed for one subscribe point.Under the hood it’s using Cloudflare Workers, queues, browser rendering, and the AI Gateway, but the TTS models aren’t very good on Cloudflare so I switched to the MAI-Voice-2 models with OpenRouter.I also created a public library you can subscribe to without ever signing up, or if you sign up, you can combine a bunch into a single feed and use your own sources as well and choose from other voice options if you don’t like “Excited Ethan” at 1.1x speed.Feed
Moozonian News
github.com• Jul 22, 2026• 1 min read
Show HN: Readmer is a compiler for building READMEs from Liquid/Jinja2 templatesAs a somewhat prolific open-source author over the past 25 years, I maintain hundreds of Rust crates, Ruby gems, Python packages, and so on. One of the painful aspects of large projects consisting of dozens of packages has been the individual READMEs for each package, with lots of repetitive elements but also a great deal of necessary package-specific variation.Now, before anyone asks, I don't want to just have an LLM generate my READMEs. People are increasingly sensitive to AI slop, and that certainly starts with a project's README: slop is readily recognizable through LLMs' lack of good judgment on README contents. More isn't always better: "If I had more time, I would have written a shorter letter" and all that.I believe that at this point given widespread, complex formatting of READMEs, what with CI badges, headers, footers, tables, images, inline HTML, etc, etc, we've in any case realistically drifted quite far from the original human-readable vision of Markdown as such. Perhaps,
Moozonian News
windowhints.com• Jul 22, 2026• 1 min read
Show HN: Window Hints – Keyboard-first window switchingHello!I’m a big fan of tools and apps that make common (preferably all) actions directly accessible from the keyboard. For example, I use Vimium every day, and it makes me so much faster. When I'm in the zone, using it feels like I'm "flying" :DAt the same time, people often use multiple displays with tons of windows open across multiple apps. Switching between these using the mouse, cmd+tab or Spotlight (by typing the app's name) kinda works, but it's easy to lose your flow when doing so.That's why I built Window Hints. It lets you jump directly to any window using only your keyboard.To get a feel for how it works, it's best to watch some demos on the website:https://windowhints.com- When you press a keyboard shortcut, Window Hints shows labels over all visible windows. Type a label to jump to the corresponding window. You can also press the shortcut again to include buried windows.- It also works when you reveal all windows with Mission Control (ctrl+↑ or swipe up with 3 fingers), as
Advertisement
Moozonian News
github.com• Jul 22, 2026• 1 min read
Show HN: SyncLite – An embedded synchronization layer for SQLite and DuckDBHi HN,We have been building SyncLite, an open-source platform to simplify offline-first application development.If your application uses SQLite or DuckDB, SyncLite lets it continue using the local database as usual while synchronization happens in the background.For SQL-based applications, SyncLite records transactions into transactionally consistent .sqlog files using SQLite as an open replication format. A background consolidator replays those transactions, derives row-level CDC and schema changes, and applies them transactionally to destination databases like PostgreSQL.The same architecture supports both database replication and many-to-one data consolidation. For applications that don't work directly with SQL, SyncLite also provides Store Devices (CRUD APIs) and Streaming Devices.SyncLite is open source and available for Java, Python, Rust, and native C/C++ ( nodejs, .NET, go in roadmap).GitHub: https://github.com/syncliteio/SyncLiteWe wouldd love your feedback on the architecture
Moozonian News
github.com• Jul 22, 2026• 1 min read
Show HN: Cactus Hybrid: We taught Gemma 4 to know when it's wrongHey HN, Henry & Roman here from Cactus.A small, on-device model is fast and private, but sometimes wrong, but frontier models are getting expensive pretty fast. So, we post-trained Gemma 4 E2B post-trained to know when it's wrong. Every response comes with a confidence score between 0 and 1. Developers can accept the on-device when it's high, hand off to a bigger cloud model when it's low. By routing only 15-35% of queries to Gemini 3.1 Flash-Lite, Gemma-4-E2B matches Gemini 3.1 Flash-Lite on most benchmarks.- ChartQA: 15-20%- LibriSpeech: 25-30%- MMBench, GigaSpeech, MMAU: 30-35%- MMLU-Pro: 45-55%We were always frustrated by the routing signals hybrid apps rely on: asking the model to rate itself in text (unreliable, and you're parsing prose), or token entropy heuristics (barely better than a coin flip in our tests). So we did mechanistic studies on small models, Gemma 4 particularly, and found the hidden state for different layers carry meaningful self-awareness signal for various si
Moozonian News
news.ycombinator.com• Jul 22, 2026• 1 min read
Ask HN: How to get started with a sustainable bootstrapped business of your own?Experienced Software Engineer working in Big Tech since the past decade. I'm at a point in my life where I'm feeling pretty done with a daily 9-5 job especially because of the overwhelming demands and random priority changes that expect you work beyond scheduled hours.I could do this in my 20s and before I had a family but I've realized that it has become increasingly hard for me lately to manage everything especially raising a family. I'd like to spend time daily with my kids and don't want to pull 40+ hour work weeks anymore.I feel like a sustainable pace would be working 3-4 hours a day and building something. I love building things so that always excites me but I'm not so excited/challenged by the people aspect of a job where you are required to be influential across orgs especially in as a Senior engineer in BigTech.Anyway, all of this is just context for what I've been thinking since long: to quit my BigTech job and build something of my own on the side that generates just enough
Moozonian News
fibertoy.dev• Jul 22, 2026• 1 min read
Show HN: I Made Shadertoy 2.0Hey everyone, about 2 years ago I was interested in making a new kind of shadertoy, and was given a lot of great ideas. After I finally had a bit of free time, I can now share what I made.Fibertoy is essentially shadertoy but you can write react-three-fiber as well as glsl code. It essentially gives a Threejs environment which means you can also just create basic 3D scenes as well as shader scenes.Here are some examples of some scenes I made with it:- Basic Voronoi scene https://www.fibertoy.dev/view/j978r67mc84yg2d36g3z5n2nqd8as1...- Cone marching example https://www.fibertoy.dev/view/j97c1kqg8n6w0st2hk3dcq7ynx87rn...- Standford bunny scene https://www.fibertoy.dev/view/j971mkwktr7bsen3ybyxks08vh87pd...As you can see from the last one, you can use your own 3D models as well. I haven't made an asset manager yet, but if you have a 3D model online, you can technically import it. I'll make an asset manager if people start using this app.I would love to get any feedback. I also have the wh
Moozonian News
rombugearth.com• Jul 22, 2026• 1 min read
Show HN: ROMbug Earth – a browser that shows where websites connectI built ROMbug Earth, a Windows browser that displays the IP address, host, estimated network location, and third-party connections behind the websites you visit.It also includes WebRTC, DNS, and IPv6 leak checks, plus an optional whole-device VPN using WireGuard. The browser and network inspection tools are free; the VPN costs $4.99 per month with no long-term contract.The goal is simple: most browsers show you the webpage, but not the network activity happening behind it. ROMbug Earth is designed to make that activity visible.Site: rombugearth.comI’d appreciate feedback on the concept, interface, and what network information would be most useful to show.
Advertisement
Moozonian News
codifycli.com• Jul 22, 2026• 1 min read
Show HN: Codify – Terraform for Developer EnvironmentsSetting up developer environments is surprisingly manual. Most teams rely on a mix of docs, shell scripts, and tribal knowledge. We built Codify to make the process reproducible with a Terraform-like workflow.The project consists of an open-source CLI, a JSON5 based configuration language, a library of 50+ resources, a web/desktop editor, and an AI assistant. The workflow mirrors Terraform: plan + apply to make changes, and import + refresh to synchronize an existing configuration with the current system. Codify manages both installation and configuration, supporting tools like Git, SSH keys, VS Code, WebStorm, Python, JavaScript, Docker, cloning git repositories and even managing macOS system settings.Compared to existing tools like Nix, our goal is to make Codify more approachable. We chose JSON5 instead of a dedicated DSL to reduce the learning curve, and Codify installs and configures packages the same way a human would, allowing it to coexist with existing installations. The trade
Moozonian News
cruciblebench.ai• Jul 22, 2026• 1 min read
Can a MUD evaluate LLMs? A $99 proof of conceptI'm the author of a paper my friends and I wrote after we were curious if a MUD, text games originating in the 1970s, could be used to evaluate LLMs. We've spent the last several months on nights and weekends running this experiment and writing the paper on just our personal computers with about $99 in API credits.Our experiment did have an interesting leaderboard but even more surprising was the measurements of each LLM. We scored each on four behavioral dimensions, two of which lean heavily on an LLM classifier. When we removed those two, one of the frontier models fell six positions. When we then checked the classifier against a second judge, the per-model agreement between them ranged from 85% to 22%. The aggregate kappa (0.04 on probe detection) indicated the instrument was noisy without saying which models the noise was hitting. The most affected model shared a model family with the classifier. This isn't proof of bias, just one observation we recorded.We realize LLM judges can b
Moozonian News
bento.page• Jul 22, 2026• 1 min read
Show HN: Bento - An entire PowerPoint in one HTML file (edit+view+data+collab)Over the past few months, our team has been building more and more slidedecks using web frontend technologies with coding harnesses like Claude Code, but a common complaint is to make even small edits we need to edit the code either manually or via the harness.To avoid this loop, I ended up creating Bento, a single HTML file with everything you need in a slide tool including animations and shared editing. There's no install or cloud login, everything works offline. The default deck is around 560 KB and it doesn't need to fetch anything once you got it.Open it in a browser and then you can edit, present, print and save. Share it via email or via Airdrop and all they need is a browser to edit, present and also do live collab on the slides. Drop it in to Claude or ChatGPT to transform existing pptx files into Bento slides. There is no cloud involved, only an encrypted blind relay to allow for shared editing. The relay doesn't see any of the data.Check it out at https://bento.page/slides/
Moozonian News
trifle.io• Jul 22, 2026• 1 min read
Show HN: Trifle – Open-source analytics that stores answers, not eventsTrifle is an open-source time-series analytics library that aggregates nested counters instead of storing raw events. All in the database you already have. After rebuilding it twice over 10 years, it now tracks ~1B events a day at my day job.It started in 2015 as my own Rails APM. I plugged into ActiveSupport::Notifications, got a few small users, and one bigger one whose scraping app broke everything. That sparked the core idea: aggregate counters into pre-defined time buckets, so a single write increments multiple buckets at once. The APM eventually faded away without much traction.Later in 2021 I needed analytics at my day job. Instead of going for something out there I revised the idea of Trifle as a more generic analytics library, borrowing some data warehouse ideas. First used Redis, then Postgres, eventually MongoDB. Hence why Trifle::Stats comes with multiple drivers that keep the DSL unified while storage layer changes with your needs. In our case (huge write volume, some read
Moozonian News
github.com• Jul 22, 2026• 1 min read
Show HN: DataParade – generate dataflow diagrams from code for risk assessmentsAs a privacy and security engineer for the past nine years, I’ve probably done close to 1000 privacy and security risk assessments. The pain points and pet peeves I’ve come across in this work led me to build DataParade. Questionnaires suck, and so does cramming in a risk assessment right before you want to launch.PARADE stands for Processing Activity Risk Assessment Diagramming Environment. The premise is simple - a dataflow diagram, embedded with security and privacy controls and metadata, can answer all the questions in a security or privacy risk assessment. Why dataflow diagrams? Because risk follows dataflow. And because a picture really is worth a thousand words, especially in this context.Early feedback told me that nobody wants to create another artifact from scratch. So my first priority has been to build out a code-scan-to-diagram pipeline which allows you to bootstrap diagrams using our CLI tool (https://github.com/DataParade-io/dataparade-cli)This is what the CLI tool does
Advertisement
Moozonian News
housecat.com• Jul 22, 2026• 1 min read
Show HN: Housecat.com – Gmail + durable workflows + sandbox VMHey we are Housecat.I've spend the last 15 years building dev tools at Heroku, Convox (YC S15), and Segment. Dev tools have gotten insanely good over the decades, now we're working to level up productivity tools for everyday work.Housecat is an email inbox running on an single-tenant agent computer with a durable workflow engine and connections to common tools for day-to-day work.My primary workflow with the product is for customer support: get a support email, trigger a triage process, open a GitHub issue, spin up a coding agent to fix it, and update the customer along the way. All from the same UI and sandbox.We've experimented with lots of different approaches to new workspace tools, and learned that many people still need a great GUI. We've landed on the email inbox as a very familiar and permanent surface and are building tools and automations directly inside it.Under the hood every user gets their own single-tenant sandbox VM and persistent SSD on exe.dev.User connections to Goog