Hnrss

Article URL: https://render.com/blog/infrastructure-patterns-for-agentic-applications Comments URL: https://news.ycombinator.com/item?id=49100035 Points: 23 # Comments: 2

Article URL: https://blog.cryptographyengineering.com/2026/07/29/some-notes-about-anthropics-new-results/ Comments URL: https://news.ycombinator.com/item?id=49099804 Points: 11 # Comments: 1

Hi HN, Rohit here from Tokenless ( https://usetokenless.com/ ), which I’m building alongside co-founders Andrew and Kev. We’re building an API gateway which routes agent traffic dynamically turn-by-turn between different models to save on AI spend. The cost of AI tokens is top-of-mind for many. Companies like Uber and Salesforce have been complaining about blowing their yearly AI spend faster than expected. Frontier models are amazing for dev work, but are so expensive. Open-source models are ch

Article URL: https://www.thefp.com/p/tyler-cowen-youll-learn-to-love-ai-writing Comments URL: https://news.ycombinator.com/item?id=49099085 Points: 1 # Comments: 0

Article URL: https://www.emergingtrajectories.com/lh/commodification-and-circularity/ Comments URL: https://news.ycombinator.com/item?id=49099069 Points: 1 # Comments: 0

There is an article sitting at 60+ points now at the homepage which is clearly written by an AI and most comments saying something along the lines "this article is clearly AI slop". In this particular case, the article seems to be real AI slop: it creates a nonsensical benchmark to compare two tools and uses that to plug the company's product. The people commenting are right to point it out. The problem with the repeated complaining about AI slop is that it makes browsing HN tiring for two reaso

Hi HN! I made Sitrep, which is an AI incident copilot that sites alongside you during incidents and watches your screen. You can chat with it, and ask questions ("what was that trace I saw?") and it uses vision to answer your questions, providing evidence. It can also integrate with existing observability tools too, I've added Grafana as an example. This means the agent can not only reference context from vision, but also verify context by using actual logs/traces. To avoid duplicate frames, it

Hi, I am a Solo developer and I have created a Chrome extension to help prevent email phishing in Gmail and Outlook. This extension tells you why an email has phishing signs and also blocks the phishing URL if it finds something suspicious. I am posting here to get feedback so I can improve my extension further. Although it's still in the early stages, I am improving it with time. If you guys could check it and give me any feedback, that would be very helpful for me. Anything good or bad is welc

Article URL: https://github.com/vincentping/claudeFS Comments URL: https://news.ycombinator.com/item?id=49098890 Points: 2 # Comments: 0

Article URL: https://www.bloomberg.com/news/articles/2026-07-29/wall-street-picks-ai-winners-and-losers-as-credit-swaps-surge Comments URL: https://news.ycombinator.com/item?id=49098861 Points: 2 # Comments: 0

I'm a director/exec, not a software engineer. Background's mostly data analytics, SQL and Excel day to day, some R and PostgreSQL. Years ago I could hack together basic HTML/CSS/PHP too. Got into AI early and now use ChatGPT and Claude constantly. With their help I've built a handful of web apps, some Python scripts and desktop tools, and an Android app on Supabase with real multiple users. So I can build things. But I'm very aware there are big holes underneath it. I can usually get something w

Article URL: https://web-ai-sdk.dev/playground/ Comments URL: https://news.ycombinator.com/item?id=49098813 Points: 2 # Comments: 1

Hey HN, I'm Oli, Software engineer building Zaatar to to automate the organic growth on my different side projects. The AI orchestrator I've built feels kind of magic tbh, I hope some of you guys will find it useful! Cheers Comments URL: https://news.ycombinator.com/item?id=49098777 Points: 2 # Comments: 0

Article URL: https://karlbode.com/ai-doomsday-bullshit-is-getting-tired/ Comments URL: https://news.ycombinator.com/item?id=49098716 Points: 8 # Comments: 4

Article URL: https://techcrunch.com/2026/07/29/as-ai-content-floods-the-internet-pangram-raises-9m-to-detect-it/ Comments URL: https://news.ycombinator.com/item?id=49098695 Points: 2 # Comments: 0

Article URL: https://github.com/hbkdad/arc Comments URL: https://news.ycombinator.com/item?id=49098638 Points: 2 # Comments: 0

Article URL: https://mergify.com/reports/state-of-merge-queues-2026 Comments URL: https://news.ycombinator.com/item?id=49098635 Points: 2 # Comments: 0

Hi HN, I built a specialized inference engine for running 4-bit Gemma 4 26B-A4B-IT on any M-series Mac using about 2 GB of RAM. It is called TurboFieldfare and is written in Swift and Metal. I have always adored on-device AI. It feels like magic that you can run a powerful NN on your Mac or iPhone. So I wanted to push the limits a bit and run a model whose weights don’t fit in memory. The model’s 4-bit quantized weights occupy roughly 14 GB, which makes running it with conventional inference too

Article URL: https://huggingface-anatomy-of-frontier-lab-model-intrusion.static.hf.space/index.html Comments URL: https://news.ycombinator.com/item?id=49098466 Points: 83 # Comments: 27

So i have been coding since last 7 years and launched multiple projects but i do lack behind in marketing , so what are the best ways to market in the current time which is organic and connects or makes a human connect with the product ? I am building curently propi.pro . Comments URL: https://news.ycombinator.com/item?id=49098454 Points: 3 # Comments: 0

Article URL: https://juliahub.com/blog/frontier-models-physical-ai-evaluation Comments URL: https://news.ycombinator.com/item?id=49098388 Points: 9 # Comments: 0

Article URL: https://github.com/AstralXVoid/NoWreck Comments URL: https://news.ycombinator.com/item?id=49098384 Points: 3 # Comments: 0

Article URL: https://www.404media.co/substackers-say-new-ai-detection-tool-is-a-witch-hunt/ Comments URL: https://news.ycombinator.com/item?id=49098357 Points: 2 # Comments: 1

Article URL: https://www.nytimes.com/2026/07/29/business/economy/data-center-electricians-training.html Comments URL: https://news.ycombinator.com/item?id=49098198 Points: 61 # Comments: 78

Article URL: https://aistack.imec-int.com/blog/gpu-self-hosting Comments URL: https://news.ycombinator.com/item?id=49098130 Points: 50 # Comments: 15

Article URL: https://arxiv.org/abs/2607.25398 Comments URL: https://news.ycombinator.com/item?id=49096969 Points: 206 # Comments: 132

Article URL: https://enklypesalt.com/posts/context-collapse-part3-ai-worming-through-word/ Comments URL: https://news.ycombinator.com/item?id=49096188 Points: 248 # Comments: 191

A TUI for Spotify, built in Rust. Uses Librespot for playback. I made it because I was bored of the spotify_player UI and wanted something cleaner. A lot of the code has been reused from that project. As for the visual style, it was inspired from Noodle (link in repo). I used an LLM to fuse the 2, and built this on Ratatui. Fortunately for me, people started using it and I was lucky enough to get contributors to help clean up and polish the project. Looking forward to feedback, thanks! :D Instal

Now that writing code is largely automated, I've seen Github being filled to the brims with slop, and it feels like the overall quality of everything produced with code (either the tooling or the software itself) has been getting worse. Pre-AI, times were a bit simpler: You looked at the stars, looked at the issues, poked around the code, and gave the repository a chance. Today there's just way too much stuff being produced. I don't see this changing in the future, so I was thinking that I'd be

Article URL: https://github.com/ilter-ai/ilter Comments URL: https://news.ycombinator.com/item?id=49093993 Points: 1 # Comments: 0

Democratisation of local AI is key. I've been working on pushing the limits of commercial hardware, squeezing any extra bit possible. My Scientific Agentic AI hareness helped me to reallocate every single bit of it. I rewrote the Kernel, I went down the CUDA rabbit hole until I have been able to explain any bit and any ms of computational power involved in the process pushing the Qwen 30B-A3B from 8 tok7s to 19 tok/s with llama.cpp up to 22.2 tok/s with my project and 109 tok/s on not novel cont

If U.S. labs slow down AGI development, this could be 2028: Moonshot Kimi 6, Alibaba Qwen 5, Z.ai, and MiniMax are all claiming AGI-level capabilities. Meanwhile, people barely talk about Anthropic or OpenAI—just like Gemini’s situation in 2026. Digital nomads are moving to Shanghai to flex cheap tokens. Major U.S. software companies are shifting core development teams to Europe and India ( Because they need to use Chinese AI) . China is supplying data-center infrastructure to developing countri

metrIQ started as a spreadsheet for my own training. The numbers were fine, but a spreadsheet never tells you anything — it just holds what you typed. So I switched to asking ChatGPT. It could explain things, but it forgot my history, answered the same question differently on different days, couldn't give me the stats I wanted, and buried it all in generic advice. I wanted a coach who remembered me. So I built my own. The first version was a rough thing in Lovable, mostly because I wanted an exc

Question is specifically for GTM engineers and performance marketers. Curious if any agentic AI ad optimization tools have actually worked for you? I tried PMax on Google Ads and our sign ups increased but pipeline didn't move (most sign ups weren't relevant audience). I am evaluating Factors and Pixis. Curious what else should I look at? Comments URL: https://news.ycombinator.com/item?id=49092716 Points: 3 # Comments: 0

I have around four for talking through ideas and three active coding agents. None of them are truly long-running. Is that typical? How many agents are you using regularly, and are any of them running autonomously for hours? Comments URL: https://news.ycombinator.com/item?id=49092442 Points: 3 # Comments: 10

For the past year I've been using VScode as my terminal space, however with the constant friction of reopening sessions, managing different projects, and re-explaining context so agents stayed useful. TrueDesk was my answer. Of course aside from it being open source: Real CLIs side by side, multi-pane layout, session restore, keybinds that smoothly transition windows, and memory/MCP automatic so you don't run a second ops stack. Clean UI that just feels good. Can't forget about the sweet rust +

Article URL: https://www.bocazasgames.com/ Comments URL: https://news.ycombinator.com/item?id=49092255 Points: 1 # Comments: 1

Grant Sanderson (3Blue1Brown) created Manim, the Python library he uses to make the math animations in his videos. We reimplemented Manim with the same Python API, but the implementation underneath is Rust, connected to Python through PyO3. The Rust code uses wgpu, so rendering happens on the GPU. To run it in the browser, we compiled the Rust parts to WebAssembly so the PyO3 extension loads in Pyodide. In the browser, wgpu targets the WebGPU API, so animations render in real time on your GPU th

The mouse was built for doing one thing at a time. One cursor that navigates a rich virtual world to accomplishing things in series. With agents my work now all happens in parallel. But the input devices we use hasn't changed yet. What I want is not more buttons and tabs in the pixel world, but some new tactile device in the real world to jump between different virtual workflows. I want feedback about all my parallel streams of work, and direct access to them from my physical world, without drag

Hi HN, Annoucing the release of Magpie, a CLI tool which turns your AI agent into an expert bookkeeper, based on Jaybase. ( https://news.ycombinator.com/item?id=48999936 ) As someone who owns a small, tech business, I have very simple books. somewhere between 10-30 transactions a month, a few depreciation line items, and a couple budgets to monitor. So it always bothered be that I was paying hundreds of dollars per year for software, and maybe a live bookeeper on top of that, just to keep these