oqoqo.ai
Show HN: Oqoqo – build evals and custom benchmarks for real-world taskshttps://oqoqo.aiMost benchmarks today exist in curated environments and do not translate well to the real world.We built Oqoqo to bridge this gap. Oqoqo makes it super simple to build realistic evals and custom benchmarks for tasks users actually care about.Oqoqo can:- Reliably measure how agent friendly your product surfaces are against Codex, Claude Code, OpenClaw, Hermes, Pi, Opencode, Cursor, GitHub Copilot - Regression test MCP, CLI, skills, SDK, and any agent facing interface (we are continuously using Oqoqo to dogfood and improve our own MCP/CLI) - Create and share custom benchmarks for how agents discover and use your product - Compare models and harnesses for domain specific tasks - See whether new versions improve agent experienceWould love to hear feedback and thoughts on what kind of evals you are running today, if you are benchmarking agent facing interfaces, and whether you have published a custom benchmark.Aug 10, 2026 9:27 PM
bing.com
YouTube Updates its Partner Programhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cdc4578284e54846f5065d4361f1f&url=https%3A%2F%2Fwww.newsshooter.com%2F2026%2F08%2F10%2Fyoutube-updates-its-partner-program%2F&c=3481101528117783372&mkt=en-usYouTube is making updates to its partner program, and it hasn't gone down well with a lot of users of the platform, who are complaining that it hurts small ...Aug 10, 2026 9:17 PM
aisloperator.com
Show HN: Fun Web page 3D visualization of laundry room statushttps://www.aisloperator.com/brickoven-estates-laundry-room/By default, this shows simulated data. It can show real data, but we don't want to hug-of-death the laundry company IoT.Click on a machine, to see how the apartment building attendant responds to people leaving their laundry for hours in high-contention machines.This was a collaboration with Claude Code. Sometimes Claude wasted time frustratingly, sometimes Claude saved lots of time. The end product is good enough, the Easter eggs were fun, and it was completed in less time than I could've done it myself.Aug 10, 2026 9:07 PM
bing.com
Can a 30-person start-up build Britain’s answer to OpenAI?http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d3caada1c47c8895034de09193616&url=https%3A%2F%2Fwww.ft.com%2Fcontent%2F02e29d7f-9e3e-4977-a5d3-ea3f2d9efd10%3Fsyn-25a6b1a6%3D1&c=7581115940074354632&mkt=en-usLondon-based Cosine has UK government backing to build a ‘sovereign AI’ model but faces far better resourced rivals ...Aug 10, 2026 9:00 PM
news.ycombinator.com
Show HN: Stagehand – the open source SDK for browser agentshttps://news.ycombinator.com/item?id=49248980Hey HN, I’m Sam, a maintainer of Stagehand, and I’m excited to show off Stagehand v4: the SDK for browser agents.For agents to do most knowledge work, they needs access to the internet. Today most people allow their agents to use Playwright to control a browser for them.But Playwright was built for testing, not agents.If you’ve ever tried to give it to your agent, you’ll quickly run into bloated context windows, inability to use iframes, and painful state sync issues (on remote browsers).We built Stagehand specifically for agents. v4 fixes all of Playwright’s issues, with better page snapshotting, nested iframe support, webmcp, self-healing primitives, and we even rebuilt it to run as an extension in the browser which minimizes round trip time for all requests.We’re roughly ~80% more token efficient and 2x faster than Playwright.Use AI primitives for self-healing automations in natural language: // Act: execute natural language actions await stagehand.act("click the login button"); // Aug 10, 2026 8:06 PM
github.com
Show HN: Gitseq: A Repo Becomes a Workroomhttps://github.com/generalbusiness-ai/gitseqLast week I was trying to figure out how to run a new project at work, where there are lots of technical artifacts, lots of stakeholders upstream and downstream, plenty of documentation for different audiences... all the usual stuff, and actually very little "code" and lots of "best practice" that the project will manage. I won't even be the subject-matter expert for most of it. Yikes.Then I figured out a small thing that could make the whole job easier: git plus a sequencer plus some ontology. Proving the idea took a few hours, until it was bootstrapped enough to use as its own development harness with a handful of agents doing most of the implementation.Work in progress - plenty still to do - but I think it's useful for agentic development. Hopefully it will eventually turn out to be the "PMO in a box" that I really need for that project.A few artifacts:- Blog post: https://generalbusiness.ai/blog/2026-08-09-gitseq/ - "Coordination and Traceability: Not Two Problems"- GitHub repo: htAug 10, 2026 7:06 PM
github.com
Show HN: Django-Style Queries for SQLAlchemyhttps://github.com/ViAchKoN/sqlalchemy-query-managerHey!Wanted to share the project I've worked on for more than a year. The core was done last summer but since then I couldn't find time to implement all the functionality I had in mind. But a few months ago, I made myself sit down and plan the introduction of the remaining things.And with the release of 0.7.0 I can say that 85 percent is done and I'm ready to share it.I wrote a Python package which serves as a layer on top of SQLAlchemy and allows you to write Django-style queries. The idea came from work where we had an additional helper file which had methods for select, filters and other simple queries. When we didn't use those helper methods we wrote queries using the SQLAlchemy syntax which took a lot of time and code space. Especially the fact that you need to specify a model in filtering.In the end I decided to move all those helper methods to a package thereby we can just import them from one place. Django style format was selected because, personally, I found it easy to read anAug 10, 2026 6:03 PM
openparser.dev
Show HN: OpenParser – swappable OCR models in one APIhttps://www.openparser.devWe build document AI workflows, and OCR is finicky. At scale, it becomes hard to debug errors and compare confidence scores across providers.New OCR models appear all the time, so we wanted one interface that lets us swap models in and out without rewriting around each provider. So far, we’ve integrated nine models from PaddleOCR, Mistral, Azure, Google and AWS Textract. We run them at cost, with no markup.We also created and open-sourced a shared schema, openparser@1, plus adapters that normalize each provider’s response. You can switch OCR models without changing the rest of your application:https://github.com/eigenpal/openparserI’d appreciate feedback on the schema and which models or providers we should add next.Aug 10, 2026 5:08 PM
bing.com
Trump Administration Bars Medicaid Funding of Transition Treatment for Transgender Youthhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d1366775b4414be438e96bec532e0&url=https%3A%2F%2Fwww.nytimes.com%2F2026%2F08%2F11%2Fus%2Fpolitics%2Ftrump-medicaid-trans-treatment.html&c=241248101916501042&mkt=en-usThe Trump administration on Tuesday barred federal Medicaid dollars from being used to provide gender-transition medication and surgery to transgender adolescents. The treatments ...Aug 10, 2026 5:00 PM
bing.com
The story of Jackie the famous Big Bear bald eaglehttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d3786bb5f488c9b6fba172ea96e43&url=https%3A%2F%2Fwww.msn.com%2Fen-us%2Fnews%2Fus%2Fthe-story-of-jackie-the-famous-big-bear-bald-eagle%2Far-AA29Of91&c=12988268447392646989&mkt=en-usJackie the Big Bear bald eagle was, arguably, the most famous bald eagle in the world. She and her partner, Shadow, gained millions of fans as they raised their chicks and battled storms in their Big ...Aug 10, 2026 4:37 PM
stoaexchange.com
Launch HN: Stoa Markets (YC S26) – A Marketplace for GPUs and AI Servershttps://www.stoaexchange.comHi HN, we’re Eren, Berat and Kaan. We’re building Stoa (https://www.stoaexchange.com), a marketplace for new and used GPUs and AI servers.GPUs are the collateral in the data center buildout. Today, financing terms mostly depend on the offtaker, meaning the company that has committed to use the compute.If that company is a hyperscaler, the financing can look investment grade. If it’s a smaller cloud or startup, terms get expensive fast, even with the same hardware as collateral.The lender’s problem is pretty reasonable. If the borrower defaults and we need to sell these servers, what can we actually get for them? There isn’t a good answer today.We started brokering GPU deals to understand why. It was much more manual than we expected. The hardware is still traded through phone calls, forwarded spreadsheets and long email threads.One week, a seller quoted us $200k for a server node and another quoted $240k for what looked like the same thing. Neither was necessarily wrong. They had diffeAug 10, 2026 4:35 PM
github.com
Show HN: My math notes for anything that touches MLhttps://github.com/publications-and-more/pubJust wanted to share this here for the person who would need it the most. Cheers :)I plan to maintain this as a hobby as well, don't expect anything big from It.Aug 10, 2026 3:43 PM
vocalslice.com
Show HN: Vocal Slice – Cut audio by selecting text, fully on-devicehttps://vocalslice.com/I make audio tools, and this one came out of the clip-pulling loop: scrub to find the right line, set the in point, set the out point, name the file, repeat.Vocal Slice transcribes your audio, then you can simply highlight the passage you want and the waveform jumps to exactly that span. Fine-tune with the handles, export a named clip.Windows and macOS.https://vocalslice.com wesley@vocalslice.comHappy to answer anything in replies, please get in touch!Aug 10, 2026 3:32 PM
bing.com
People Share Life-Saving Advices That Are Worth Memorizinghttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d867843ba486e88849d41a4a4dca3&url=https%3A%2F%2Fwww.msn.com%2Fen-us%2Fhealth%2Fgeneral%2Fpeople-share-life-saving-advices-that-are-worth-memorizing%2Far-AA29NNBq&c=4754775040305037807&mkt=en-usWhen danger strikes out of nowhere, it’s quite hard for us to channel our inner Hulk or Wonder Woman to save the day. Most of the time, we either freeze, panic, or do something totally helpless. For ...Aug 10, 2026 2:37 PM
bing.com
Bronny James faces pivotal Lakers season without LeBron on rosterhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d9f00a0994135a32cc5a617f572f8&url=https%3A%2F%2Fwww.jsonline.com%2Fstory%2Fsports%2Fnba%2Flakers%2F2026%2F08%2F10%2Fbronny-james-lakers-development-stats-lebron%2F91244188007%2F&c=9546325272291122937&mkt=en-usBronny James comes into Year 3 of his NBA career with the specter of LeBron James' shadow looming. He will now be the lone James on the Los Angeles Lakers roster and enters a pivotal season when ...Aug 10, 2026 1:19 PM
app.edgi.tv
Show HN: Edgi – Letterboxd or Strava for your brainhttps://app.edgi.tv/Hi HNExcited to share edgi.tv - an educational media platform where the things you learn accumulate into a visual map of your intellectual interests. The experiment is whether you can build a consumer product around intellectual identity, something like Letterboxd or Strava but for your brain.The main entry point is a curated TikTok-style feed that surfaces vids from creators like Kurzgezagt, VSauce, AstroAlexandra, Cleo Abram, 3Blue1Brown, etymologynerd, etc with interactive games accessible on top of the videos. You can also create podcasts, quizzes, etc with AI on any topic you’re interested in and earn XP towards concepts.Concepts are the unit that connect all media together and create the knowledge graph and power the rec algo. As you engage with a concept, you have a chance to earn a Concept Card (a Pokemon-style collectible for that idea). The concepts then accumulate on a mind map, which doubles as a social surface so you can see what friends are interested in and where your inAug 10, 2026 1:14 PM
rfi.fr
Days of heavy rain leave at least 16 dead in Philippineshttps://www.rfi.fr/en/international-news/20260810-days-of-heavy-rain-leave-at-least-13-dead-in-philippinesFloods, landslides and accidents driven by a nine-day downpour equivalent to more than a month's worth of rain in some areas have killed at least 16 people in the northern Philippines, rescuers said…Aug 10, 2026 10:58 AM
news.ycombinator.com
Ask HN: Is Traditional SaaS Dying?https://news.ycombinator.com/item?id=49242046I think a good proxy for the question is: Does your employer try to replicate some SaaS features in-house? And do these experiments mostly succeed?My data points: Two internal projects to replace Hubspot and Float. Hubspot will be kept, Float will be replaced.Aug 10, 2026 10:55 AM
news.ycombinator.com
Show HN: Game Engine with its own IDE and scripting languagehttps://news.ycombinator.com/item?id=49242000GitHub repo: https://github.com/ArcadeMakerSources/ArcadeMaker (first of all, i want to say that I made this project for fun and challenge. I always wanted to create my own language. This has no business goal or something.) So the project is a 2D game engine (MonoGame backend) inspired by GameMaker 8. It includes its own IDE (built with WinForms) and an interpreted programming language that I wrote myself. The language—definitely the biggest challenge in the project—is a simple dynamically typed language. When I started, I had zero knowledge of how to build something like this. I didn’t even know I was making an interpreter; at first I called it a compiler. It was a personal challenge, and I wanted to figure everything out without using any resources or tutorials. My mindset was basically: “I need to write software that takes a text file containing code and just does what it says.” Somehow, I made it work. In the beginning, running an empty loop counting to 1M took 7 seconds. After a lAug 10, 2026 10:50 AM
bing.com
How mortgage preapproval works: A beginner's guidehttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d65d451e946278cfb6c7912c019c5&url=https%3A%2F%2Fwww.msn.com%2Fen-us%2Fmoney%2Frealestate%2Fhow-mortgage-preapproval-works-a-beginners-guide%2Far-AA29MSbK&c=14993594639928548445&mkt=en-usLearn how mortgage preapproval works, what lenders look for, which documents you'll need and how preapproval can help you buy a home with confidence.Aug 10, 2026 10:42 AM
