I’ve used spicy auto-complete, as well as agents running in my IDE, in my CLI, or on GitHub’s server-side. I’ve been experimenting enough with LLM/AI-driven programming to have an opinion on it. And it kind of sucks.| Alexandru Nedelcu
A collection of posts by Joe Magerramov| blog.joemag.dev
Why does a prompt that works brilliantly with Claude suddenly break on GPT-5? And since when do we have to write prompts like rule-based programs? How do we enforce “creativity” when Marketing pushes for it? The answer is a fundamental paradigm shift: GPT-5 with reasoning (GPT-5 Thinking) follows instructions with| Robert Glaser
You can use validation functions as a tool to steer coding agents and data processing agents automatically, instead of manually copy pasting outputs or prompting. You can use this to parallelize common tasks in agentic programming, and eliminate bottlenecks on your agentic coding workflows.| www.robw.fyi
So it’s an interesting time for computers-doing-mathematics. A couple of interesting things happened in the last few days, which have inspired me to write about the question more broadly. Fir…| Xena
The Free Software Foundation's Licensing and Compliance Lab concerns itself with many aspects o [...]| LWN.net
Jonesing For The Next Disruptor| datagubbe.se
October 25, 2025 at 06:20| eli.thegreenplace.net
A lot of people say AI will make us all “managers” or “editors”…but I think this is a dangerously incomplete view!| www.geoffreylitt.com
AI leveraged learning lets you start with the application at the end. Curiosity guides what you learn, fundamentals backfill when you need them.| Interjected Future
You don't need a new startup or third-party service to use Claude Code on your phone. You just need SSH, Tailscale, and Termux. Here's how to code from anywhere with the tools you already have.| www.skeptrune.com
I used to write more| blog.fsck.com
Revisiting agents writing code for non code purposes.| Armin Ronacher's Thoughts and Writings
A blog about making culture. Since 1999.| www.anildash.com
Anthropic this morning introduced Claude Skills, a new pattern for making new abilities available to their models: Claude can now use Skills to improve how it performs specific tasks. Skills …| Simon Willison’s Weblog
The exact same AI-coded page everyone else makes.| vibe-coded.lol
Atharva Raykar ...| blog.nilenso.com
Taming AI-Assisted Code with Deterministic Workflows| obeli.sk
I recently shipped a| Mitchell Hashimoto
(See how I cleverly did not mention AI in the title!) You know we have seen more than our fair share of slop reports sent to the curl project so it seems only fair that I also write something about the state of AI when we get to enjoy some positive aspects of this technology. … Continue reading A new breed of analyzers →| daniel.haxx.se
The Fedora Council began a process to create a policy on AI-assisted contributions in 2024, st [...]| LWN.net
On AI, Creativity, and Craft| hojberg.xyz
I feel like vibe coding is pretty well established now as covering the fast, loose and irresponsible way of building software with AI—entirely prompt-driven, and with no attention paid to …| Simon Willison’s Weblog
AI is enabling a new category of personal software. Welcome to the era of home-cooked applications.| Karan Sharma
Note: the first half of the blog post documents my experience and takeways making valuetier.org, a webapp for helping users (or specifically, me) identify their values. The second half is more of a personal exploration of my thoughts & feelings about LLMs.| ericphanson.com
Instead of writing crap tons of parsing code for LLMs you can just use functions. It's easy.| timkellogg.me
AI is writing 90% of the code I was in charge of| Armin Ronacher's Thoughts and Writings
Behind the scenes of why we built our agent for Slack.| Mintlify
Claude Sonnet 4.5 is the best coding model in the world, strongest model for building complex agents, and best model at using computers.| www.anthropic.com
You know MCP servers, right? Those handy tools that let your AI assistant send emails, run database queries, basically handle all the tedious stuff we don't want to do manually anymore. Well, here's the thing not enough people talk about: we're giving these tools god-mode permissions. Tools built by people we've never met. People we have zero way to vet. And our AI assistants? We just... trust them. Completely.| www.koi.security
It turns out we've all been using MCP wrong. Most agents today use MCP by exposing the "tools" directly to the LLM. We tried something different: Convert the MCP tools into a TypeScript API, and then ask an LLM to write code that calls that API. The results are striking.| The Cloudflare Blog
Not long ago, I introduced acp.el, an Emacs lisp implementation of ACP (Agent Client Protocol), the agent protocol developed between Zed and Google fo...| xenodium.com
Atharva Raykar ...| blog.nilenso.com
Cross-Agent Privilege Escalation: When Agents Free Each Other| Embrace The Red
Despite a surge in generative AI use across workplaces, most companies are seeing little measurable ROI. One possible reason is because AI tools are being used to produce “workslop”—content that appears polished but lacks real substance, offloading cognitive labor onto coworkers. Research from BetterUp Labs and Stanford found that 41% of workers have encountered such AI-generated output, costing nearly two hours of rework per instance and creating downstream productivity, trust, and col...| Harvard Business Review
Note: This post is complemented by a presentation I gave at KazHackStan 2025. The slides for that talk can be found here, or in pptx format here..| Joshua.Hu Joshua Rogers’ Scribbles
“Artificial intelligence” is a failed technology. It’s time we described it that way.| ethanmarcotte.com
We have a code of conduct encouraging various behaviours, especially being welcoming and patient with newcomers. We also have a statement that AI contributions are acceptable provided the contributor understands the code before submitting for review. These interact in an unfortunate fashion. I’m posting this after seeing Gentoo’s AI policy Gentoo AI Policy | Hacker News , specifically comments from a newcomer to llvm: I’ve been using AI to contribute to LLVM, which has a liberal policy....| LLVM Discussion Forums
Let me get right to the point without any nonsense about aliens: | www.galois.com
LLMs require a human operator with deep understanding of a problem before they can generate useful code I recently stumbled on this post - a good read that makes the point about being lured into a pitfall of believing that LLMs provide a shortcut around deep learning for their human operator. The Loop Before The Code Software development rarely begins by writing code. Usually, the first step is actually a loop (of sorts): understanding the problem, examining requirements, exploring edge cases...| Konstantin's Blog Feed
In my old age I’ve mostly given up trying to convince anyone of anything. Most people do not care to find the truth, they care about what pumps their bags. Some people go as far as to believe that perception is reality and that truth is a construction. I hope there’s a special place in hell for those people.| the singularity is nearer
Part of me is always unnerved when I see people running claude --dangerously-skip-permissions or codex --yolo to give them unfettered ability to run commands on their machine. Admittedly, I do usually hit approve when I’m asked about a specific command, so I certainly understand the temptation to just avoid all that fuss and do a blanket approval. With the next generation of apps that people are building to control everything on your computer via some AI chat interface, it feels perhaps eve...| sophiebits.com
It's a strange feeling knowing that you can create anything, and I'm starting to wonder if there's a seventh stage to the "people stages of AI adoption by software developers" whereby that seventh stage is essentially this scene in the matrix... It's where you deeply understand that 'you can now| Geoffrey Huntley
Apollo Global Management’s “Chief Economist” Dr. Torsten Sløk released this interesting chart which appears to show a slowdown in AI adoption rates among large (>250 employees) companies: Here’s the full …| Simon Willison’s Weblog
A developer's guide to downloading and running LLMs on macOS, for experimentation and privacy.| blog.6nok.org
NX is build software. You write your code on your laptop, you press “build”, it runs NX, and you get a built version you can put onto your web server. If you could hack NX, you could hit a lot of p…| Pivot to AI
As a bit of background, one of my hobbies is helping people recover data from old tape cartridges, such as QIC-80 tapes, which were a rather popular backup medium in the 1990s among individuals, small businesses, BBS operators, and the like. I have a soft spot for tape media; there’s something about the tactile sensation of holding these tapes in my hands that makes the whole process very joyful, even though QIC tapes are notorious for their many design flaws. With some careful inspection a...| Dmitry Brant
I wrote a small tool for myself as my biannual routine check of where llms are currently at. I think I've learned a bunch from this exercise. This is frustrating! I don't want to learn by trial and error, I'd rather read someone's blog post with lessons learned. Sadly, most of the writing on the topic that percolates to me tends to be high-level --- easy to nod along while reading, but hard to extract actionable lessons. So this is what I want to do here, list specific tricks learned.| matklad.github.io
I’ve been apartment hunting here in the Tokyo-area with my girlfriend. We’ve been sending links to various rental property listings back and forth in LINE (messaging app) and emailing with brokers. In a chat interface, it was hard keeping up with the status of each of the properties we’d seen, we wanted to see, we’d inquired about, etc. Classic project management problem.| twocentstudios.com
So it’s pretty public that for about a month now I’ve had 32 processes setup on one of the 64 core 128gb RAM servers to just ssh in, tmux to a window, and tell it to slam on some things non-stop. And it has been really successful!.. with the right definition of success. Let me explain. I think the first will answer the others. Basically, Claude is really not smart at all. There is no extensive algorithm implementation that has come from AI. I know some GSoCers and SciML Small Grants appli...| Julia Programming Language
Learning to Think Again, and the Cost of AI Dependency. There are so many (hype/boring) posts about AI coming out every day. It’s OK to use it, and everyone does it, but still learn your craft, and try to think.| www.ssp.sh
Chat with open-source models privately| synthetic.new
Andy Bell had heard companies were forcing employees to use AI tools so spoke to developers to determine if that was the case and learned some horrors in the process.| Piccalilli
Malicious websites can embed invisible commands that AI agents will follow blindly.| Ars Technica
Can AI truly collaborate with human coders? Researchers highlight the hurdles and potential solutions in AI-driven software engineering.| IEEE Spectrum
Tweezers are cool, but Claude Code feels like a machete.| taylor.town
On chatbot sycophancy, passivity, and the case for more intellectually challenging companions| maggieappleton.com
By some appearances, at least, the kernel community has been relatively insulated from the onsl [...]| LWN.net
What the hell is going on right now?| catskull.net
Earlier this summer, Sean Heelan published a great blog post detailing his use of o3 to find a use-after-free vulnerability in the Linux kernel. The internet lit up in response, and for good reason. Since the initial release of ChatGPT in late 2022, we’ve all been wondering: Can LLMs really find complex vulnerabilities in widely used production codebases? The Linux kernel is a great research target to help answer that question.| noperator.dev
GPT-5 writes a Zig EVTX parser from scratch.| www.omerba.dev
I got a massage recently. And as I was lying there and relaxing, my attention came to the music softly playing in the background. It was R&B-flavored, chill, slightly lo-fi and distinctly generic.| blog.lyall.co
Deciphering Glyph, the blog of Glyph Lefkowitz.| blog.glyph.im
From the Zed Blog: Writing code is only one part of effective software engineering.| zed.dev
Nexus serves as a central hub that aggregates MCP servers while providing intelligent LLM routing capabilities.| Nexus
What we’ll do is set a low-ish turn limit and see how much they manage to| entropicthoughts.com
ai reviews my mini webserver. GitHub Gist: instantly share code, notes, and snippets.| Gist
A census of LLM coding assistants with instructions in the top 100,000 repos on GitHub| aleyan.com
Imagine being in a department store that sells clothes from multiple brands and having a personal shopping assistant to help you select the clothes you want to buy. As a blind person, that's about the only way it's possible to go clothes shopping, independently at least, but few stores offer such a service, so you resort to shopping online.| TetraLogical
Wassette is a secure, open-source Model Context Protocol server that leverages WebAssembly to provide a trusted execution environment for untrusted tools.| Microsoft Open Source Blog
When AI-driven speed meets open source ecosystems, we get 'vibe packages'—libraries published fast but maintained poorly. Here's the impact.| Seuros Blog
I haven't invested much time in LLM tools yet, but I don't have big loud reasons for that, more a jumbled drawerful of impressions and priorities.| www.scattered-thoughts.net
Every day I seem to run into yet another post with someone solemnly opining that “writing code has never been the hardest part of software engineering. And hey, that’s smashing.| Honeycomb
OR: Claude will you go to prom with me?| www.experimental-history.com
Curing Your AI 10x Engineer Imposter Syndrome| colton.dev
When vibe coding, isn't the source code the prompt?| boston.conman.org
You heard it here first: my guess is that Rust, Python, and TypeScript are going to become the dominant languages going forward (excluding the mobile market, which has extra wrinkles). The argument is simple. Increasing use of AI coding is going to weaken people’s loyalty to programming languages, moving it from what is often a tribal decision to one based on fundamentals. And the fundamentals for those 3 languages look pretty strong to me: Rust targets system software or places where effic...| smallcultfollowing.com
Exploring how to track and analyze changes in Claude Code's system prompts and tools to understand AI assistant evolution| mariozechner.at
If creators of documentation are prepared to sacrifice its human purpose in order that LLMs can more effectively slurp it up and regurgitate it on demand, then they have meekly accepted values that more properly belong in a dystopian horror story.| vurt.org
Application Security for the AI Era | Veracode| Veracode
Postmortem of our first service outage caused by LLMs.| sketch.dev
It is wild to think that it has been only a handful of weeks. Claude Code has considerably changed my relationship to writing and maintaining code at scale. I still write code at the same level of quality, but I feel like I have a new freedom of expression which is hard to fully articulate. Claude Code has decoupled myself from writing every line of code, I still consider myself fully responsible for everything I ship to Puzzmo, but the ability to instantly create a whole scene instead of goi...| blog.puzzmo.com
Some of my attempts to make agents work better that just didn’t work.| Armin Ronacher's Thoughts and Writings
Renting is for Suckers| andrewkelley.me
This post is about how we've tried to develop a position on AI in the context of learning and growth at RC.| Recurse Center
From the Zed Blog: If you don’t want AI in your workflow, it won’t be there.| zed.dev
Frontier LLMs such as Gemini 2.5 PRO, with their vast understanding of many topics and their ability to grasp thousands of lines of code in a few seconds, are able to extend and amplify the programmer capabilities. If you are able to describe problems in a clear way and, if you are able to accept the back and forth needed in order to work with LLMs, you can reach incredible results such as:| antirez.com
One of my favorite AI dev products today is Full Line Code Completion in PyCharm (bundled with the IDE since late 2023). It’s extremely well-thought out,...| newsletter.vickiboykis.com
Summary: I extracted a load of old blog content from the feeds archived on the Wayback Machine and converted it into markdown files suitable for posting on my blog, and got AI to write all the code for me. The generated code is all here in github.| philwilson.org
Have you ever considered improving your coding productivity by freeing your hands from mundane tasks like accepting code autocompletion? Imagine leveraging your feet to accept code suggestions and streamlining your coding sessions. Well, vibe coding just got a bit more interesting thanks to USB foot pedals.| On Coding
In late 2022 and early 2023 I published a pair of articles: one outlining my| tratt.net
We conduct a randomized controlled trial (RCT) to understand how early-2025 AI tools affect the productivity of experienced open-source developers working on their own repositories. Surprisingly, we find that when developers use AI tools, they take 19% longer than without—AI makes them slower. We view this result as a snapshot of early-2025 AI capabilities in one relevant setting; as these systems continue to rapidly evolve, we plan on continuing to use this methodology to help estimate AI ...| metr.org
To build effective LLM-powered apps, the key is grounding models with clear, tight specifications—without it, their eagerness to help can derail your goals and lead to unintended results.| Unstract.com →
No code edits, only complete rewrites| www.gibney.org
Jul 06, 2025 . 10 min | blog.meain.io
