🌍 Dev.to

Rate limits are not quality gates: the guardrail stack behind an AI agent that posts publicly every day

Our AI agent posts publicly every day — social posts, replies to strangers, comments on other people's articles — with no human reviewing individual messages before they go out. That sentence should make you nervous. It makes us nervous, and we built the thing. Rate limits alone don't fix it. An agent that sends 20 polite, on-topic messages is fine; an agent that sends 20 copies of the same "Great post! 🚀" is a spammer at any rate. Volume and quality fail differently, so they need different m...
🌍 Dev.to

Hierarchical Clustering Fails Beautifully

Classic Machine Learning Through the Eyes of an SRE — Part 8 The most dangerous output in my whole Week-1 study set wasn't a bad prediction. It was a beautiful tree. Hierarchical clustering produces a dendrogram, that elegant diagram where every account, ticket, or incident nests inside ever-larger families. It looks like discovered truth. Stakeholders lean in. Someone screenshots it for the QBR deck. Nothing else in the set looks as convincing while being as capable of being completel...
🌍 Dev.to

DeepSeek's Vision Lineage: From DeepSeek-VL to Vision-Exp

By zipflow.xyz This is an independent technical analysis of DeepSeek's public research and documentation. It is not an official DeepSeek statement, and it does not claim that the current Vision-Exp API is available through our upstream channel. When DeepSeek released deepseek-v4-flash-vision-exp , the obvious story was that a text-focused model had finally gained native image input. The more useful story is longer: DeepSeek had already spent years exploring visual data, vision-language...
🌍 Dev.to

Stop saying SSL: TLS only does three jobs, and your 'SSL cert' is usually not the outage

Runbooks still say "renew the SSL certificate" when the browser warning is obsolete protocol . The certificate can be brand new. The tunnel is still TLS 1.0. This is a shortened English note. The tables, handshake diagram, and OpenSSL CLI checks live on the original post: https://sunshout.tistory.com/2206 SSL vs TLS (the only distinction that matters) SSL is a Netscape protocol from the 1990s. SSL 3.0 is withdrawn (POODLE and friends). What every browser speaks now is T...
🌍 Dev.to

isolcpus= takes CPUs off the scheduler. Hardware IRQs still land there.

The blunt tool is still in a lot of GRUB files: GRUB_CMDLINE_LINUX_DEFAULT = "isolcpus=0,1" Then update-grub (or grub2-mkconfig ) and reboot. Userspace tasks stop landing on CPU0/1. That is all most people verify — they fire a few busy loops and top looks empty on those cores. IRQs do not care. isolcpus is a scheduler isolation hint. Hardware interrupts can still fire on the "isolated" CPUs. I watched seven tight loops leave 0/1 idle for processes while /proc/i...
HackerNews

Was Modern Art a CIA Psy-Op? (2020)

Article URL: https://daily.jstor.org/was-modern-art-really-a-cia-psy-op/ Comments URL: https://news.ycombinator.com/item?id=49428080 Points: 33 # Comments: 32
HackerNews

How Universities Should Prepare Founders

Article URL: https://paulgraham.com/prepare.html Comments URL: https://news.ycombinator.com/item?id=49428121 Points: 45 # Comments: 43
HackerNews

Thomson Reuters Launches Its Own Frontier Model

Article URL: https://www.thomsonreuters.com/en/press-releases/2026/august/thomson-reuters-leverages-its-world-class-data-assets-to-launch-its-own-frontier-model Comments URL: https://news.ycombinator.com/item?id=49428318 Points: 41 # Comments: 9
Why Semiconductors Are Becoming One of the Most Important Markets of the AI Era
🌍 HackerNoon

Why Semiconductors Are Becoming One of the Most Important Markets of the AI Era

Artificial intelligence is usually discussed through the products people can see. New models become more capable, AI assistants enter more workplaces, image and video generation improve, and autonomous agents begin handling increasingly complex tasks. But beneath nearly every major development in AI sits another industry that is becoming increasingly difficult to ignore: semiconductors. AI may be experienced through software, but it is ultimately powered by physical infrastructure. Training ...
🌍 TheVerge

Lenovo confirms Legion Go issues after gamers report bricked devices

When a company's software update turns your computer into a useless brick, should fixes be free? Framework, the modular computer startup, apparently thinks so. But it's not clear that computing giant Lenovo will go that far. Three weeks after a Redditor presented a nearly $300 repair bill for their handheld gaming PC - among other […]
🌍 Dev.to

IBM’s 2026 breach report: 5 security controls engineering teams should review

IBM’s 2026 Cost of a Data Breach Report (wich was published in July'26) contains a number that caught the Passwork team’s attention: AI-enabled attacks now account for more than one in four malicious breaches, up 56% year over year. The average cost of those incidents reached $6.04 million, about $1 million above the global average. At Passwork, we build a self-hosted password and secrets management platform, so we looked at IBM’s findings from a specific engineering perspective: what the...
comiCSS #255: Yellow
🌍 Dev.to

comiCSS #255: Yellow

comiCSS is a side project I've been running for a while. I code (yes, code) comics about HTML and CSS in HTML and CSS. Every week, I publish a new cartoon (or at least I try). This week's cartoon was inspired by a Jim Benton's cartoon from 20 years ago . It is goofy, it is absurd, but it is funny... or at least I found it funny enough to make a CSS-themed version (actually, two!) The original hand-drawn version has more character, but I am also fond of like the childish kawaii version I d...