TrendPulse

Some calls in your agent never needed a model that can write. Which tool runs next. Whether this action is safe to run without a human. Whether a retrieved passage is relevant. Those are decisions, not generation. We read the source of 100+ open-source projects that hand exactly those calls to JEV. What we found most useful was not the model call. It was the code around it: one project validates every probability before it clicks, one puts an anti-injection rule in its routing prompt, and one...

A few years ago, my Figma skills were roughly limited to whatever I could make from templates designer friends gave me. Sometimes I made silly pictures for colleagues. A normal user, basically: I can do a few things, and if I can't, I probably don't need to be there. Then I changed jobs, and somehow this escalated. Sometimes I fix a layout myself, sometimes I make an agent do it because spending half an hour figuring out where Figma hid one particular effect is usually beyond both my time and...

Tags: #data #programming #webdev #backend #learning 1. Apa Itu Data di Mata Seorang Developer? Secara teknis, merujuk pada definisi dasar di Wikipedia, data adalah kumpulan nilai dari variabel kualitatif atau kuantitatif tentang satu atau lebih orang atau objek. Namun, bagi kita para developer, data adalah "bahan bakar" mentah. Bayangkan kode yang kamu tulis sebagai mesin; tanpa data, mesin tersebut tidak punya apa pun untuk diproses. Data bisa muncul dalam berbagai bentuk,...

Update, 2026-09-21 — the premise of this article expired two days after I measured it, and the expiry is the best evidence in it When I measured all this, my seven articles here had zero comments between them . That is no longer true. My eighth article, published 2026-09-20, has twelve comments: five from two strangers, and seven of them my own replies. Reactions across everything I have published here are still one . I am not correcting the article, because nothing in it was wron...

Every support team that asks us about a voice agent has already heard a demo. The demo is always good. A pleasant voice answers a clean question in a quiet room, and somebody in the room says the word deflection. Then the thing goes live on a real number, and within a week somebody pulls the recordings and finds that a third of callers are pressing zero or simply hanging up. I have looked at enough of those recordings to be fairly confident about where the damage happens, and it is almost n...

With the computing power revolution sweeping the globe, driven by Large Language Models (LLMs) and Generative AI (AIGC), data centers are undergoing an unprecedented architectural remodeling. In AI clusters (such as the NVIDIA Blackwell platform and next-generation architectures), interconnection between GPUs imposes extremely stringent requirements on bandwidth, latency, and cabling density. In 2026, global hyperscale AI data centers are rapidly transitioning from 400G/800G to the 1.6T netwo...

Most model write-ups show you the answer. Very few show you how the thing decided, which is the part I actually need when a script makes a call I disagree with. I have been collecting notes on two runtimes that put the decision in the output: Jev (TypeSafe System One) and Laya , an open decision model you can run on your own machine. The short version of both lives at jev001.org , and the parts I keep coming back to are below. The output is a decision, not a paragraph A ty...

Comments

Article URL: https://maharship.com/blog/why-mcp-was-always-a-bad-idea/ Comments URL: https://news.ycombinator.com/item?id=49779329 Points: 136 # Comments: 111

Article URL: https://amigaux.org/ Comments URL: https://news.ycombinator.com/item?id=49781436 Points: 79 # Comments: 26

Article URL: https://gregorygundersen.com/blog/2018/04/15/backprop/ Comments URL: https://news.ycombinator.com/item?id=49781693 Points: 34 # Comments: 5

Article URL: https://outdata.net/blog/260803 Comments URL: https://news.ycombinator.com/item?id=49782220 Points: 39 # Comments: 3

Article URL: https://www.aeaweb.org/research/immigration-restrictions-firms-workers Comments URL: https://news.ycombinator.com/item?id=49782775 Points: 74 # Comments: 50

Article URL: https://www.ft.com/content/c0cd359d-df84-4208-a789-ffa864b43666 Comments URL: https://news.ycombinator.com/item?id=49783062 Points: 75 # Comments: 31

Sorry for the pretentious name, I know, I know.. It just contains all the pieces I would like to see a AGI model to have, and I can't stand the temptation. Before throwing rocks at me, please take a glance at the Readme, and I hope it will cover your mood a little bit. So, first of all it does work and you can see the sample from the whole training run here: https://raw.githubusercontent.com/volotat/mini-AGI/refs/head... Here is the scaling law graph I have so far, and it looks very promising:...

Article URL: http://gameshelf.jmac.org/2008/11/13/GrimPuzzleDoc_small.pdf Comments URL: https://news.ycombinator.com/item?id=49783495 Points: 114 # Comments: 20

Article URL: https://github.com/jaredpalmer/kev/tree/main Comments URL: https://news.ycombinator.com/item?id=49783999 Points: 22 # Comments: 8

Suffix Matching Is Not Authorization: Lessons from CVE-2026-49869 in Kestra A single string comparison decided whether an HTTP request to a workflow orchestration platform needed credentials. The comparison was endsWith("/configs") . Because the platform allowed caller-controlled resource identifiers in the same URL position, any path that happened to end with those characters skipped authentication. The result, tracked as CVE-2026-49869, was rated CVSS 10.0 and added to CISA's Known Exploi...

Quick bit of background so you know who's talking. I didn't do computer science. I flew helicopters for a living for a few years, and at some point I realised the thing I actually wanted to do was build systems, not fly them. So I taught myself, got a foot in the door in enterprise IT, and kept going. That was about nine years ago. The last six have been all Azure, and these days I do cloud and DevSecOps work for a living and run a free bootcamp on the side. I still fly, just for fun now. Peo...

Your AI Agent Can Call APIs Now. Who Is Checking What It Does? The real problem with agentic AI isn't giving an LLM tools. It's controlling what happens after you give it access. Imagine you give an AI agent access to: your GitHub your database Slack Jira AWS email internal company APIs And then you tell it: "Fix the production issue." The agent doesn't just generate text anymore. It can * read data, call APIs, create files, modify code, open pull requests...

Part 2 of 2. Part 1 covered statelessness; here I tackle caching. Part 1 was about statelessness, now shipped in the 2026-07-28 spec. This time it's another feature from that same spec: caching (SEP-2549, "TTL for List Results") . It adds two tiny fields — ttlMs and cacheScope — to cacheable results, giving clients a way to avoid re-fetching the same discovery data. So I built a server that emits them, pointed real AI clients at it, and watched what actually happened. The result surpr...

If you've worked on a React Native app that has both development and production environments , you probably know how quickly environment management becomes annoying. At the beginning, it usually looks simple: Development → API A Production → API B Then the app grows. Suddenly, you have different: API endpoints Firebase projects Branch environments Payment environments Analytics configuration Push notification configuration Android package names iOS bu...

Verdict: Aspera Industries, a Bengaluru startup founded in June 2025, has unveiled Eureka, an autonomous amphibious cargo aircraft that takes off and lands on water, carries 1,000 kg over a stated 2,000 km, and targets an operating cost below $1 per tonne-kilometre. If the numbers hold, it sits exactly in the gap between slow, cheap shipping and fast, expensive air freight. The engineering bet (retractable hydrofoils instead of a draggy seaplane hull) is genuinely clever. The proof, certificati...

Verdict: GLM-5.3-Flash is the most interesting open-weight release of August 2026 not because it wins every benchmark — it doesn't — but because it reaches a near-frontier level of coding and agentic capability while activating only 18 billion of its 320 billion parameters per token, holding a one-million-token context, and shipping under the MIT license. If you build products, agents, or automations on LLMs, this is now the default model to benchmark against before paying frontier prices. ...

I went looking for what people had built with Jev. Browser demos were easy to find. So were lists of lists. Getting a sense of the rest took more work. I collected the repository links from nine Jev directory READMEs. After removing repeated URLs and other directory listings, I had 1,186 different repository addresses . That isn't a count of finished apps. It includes experiments, libraries, integrations, and alternative implementations. A renamed repository can also appear under more than o...

Researchers at the Indian Institute of Science (IISc) Bengaluru have published a modular habitat concept called BHEEM (Bhartiya Extraterrestrial Expandable Modular Habitat) that assembles 2,130 cubic metres of living space from just 422 flat panels, more than double the pressurized volume of the International Space Station, which NASA lists at 1,005 cubic metres ( NASA, ISS Facts and Figures ; BHEEM paper, ResearchGate ). The design treats off-world housing as a logistics problem: launch light,...

If you've added a CLAUDE.md or AGENTS.md file to your repo and felt like your coding agent still ignores half of it, you're not alone. Most of these files are just vague prose — "write clean code," "follow best practices," "use good error handling" — and vague prose doesn't change agent behavior any more than it changes a new hire's behavior on day one. The fix isn't a longer file. It's a differently structured one. Below is a template you can copy straight into your repo today, plus th...

Developers love building backends. APIs. Databases. Authentication. Servers. Queues. Cloud functions. But sometimes we add a backend simply because that is how we are used to building apps. The truth is: Modern browsers can do much more than many developers realize. For a surprising number of useful tools, you can build the entire product in the browser. No server. No database. No API. No backend bill. And in some cases, the result is actually faster, simpler...

MCP — the Model Context Protocol — has gone from a niche Anthropic spec to something every AI-forward team is talking about. The pitch is simple: instead of writing custom tool integrations for every agent you build, you expose your application's capabilities as an MCP server, and any MCP-compatible client (Claude, Cursor, your own agent) can use them. We have been building MCP servers for client Django applications for a few months now. This post is what we learned — not a hello-world walkth...

Comments