This post is about prompt-based command and control (C2), which is becoming more relevant. What is Promptware-Powered C2? Three years ago, when ChatGPT introduced the browsing tool, we already…

All you need is Agents and Human intent Took 3rd place (according to the AI judge) and won the "Scariest" award at the NEBULA:FOG, an AI × Security hackathon in San Francisco on March 14, 2026. My…

I’ve been very critical of AI but have never really used it in depth and I feel that needs to change. Don’t criticize what you don’t know. I’m going to ignore the moral,…

¶ ## SE Gyges, Mar 16, 2026 ¶ [link to article](https://www.verysane.ai/p/polly-wants-a-better-argument?utm_source=activity_item) ¶ this gave me a lot to think about. i think ill wana write more…

The original tool relies on sweet-cookie Phase 1

I should probably just be making individual posts for these rather than trying to wait and gather enough links to make a round-up with (or stick to two or three at most maybe). Been sitting on some…

How does Claude Code's new /btw command work, figuring it out from the minified source.

A Sydney tech entrepreneur, Paul Conyngham , had a rescue dog, Rosie, with advanced mast cell cancer and months to live after standard treatment. He paid for Rosie's tumor sequencing through the…

There is a recurrent question when starting to use LLMs seriously : “How do I structure my prompts ?”.

In two earlier posts I covered a bit of the history of the current batch of AI models, what they are good at, and what they’re not so good at. Had I published those posts a year earlier, we probably…

This is pretty exciting. I heard about Kaparthy's Autoresearch only vaguely: it's a way of getting the LLM to run experiments on itself(?), running for long periods of time. It wasn't until I read…

A practical guide to maximising engineering productivity with AI tooling — covering context management, instructions files, testing, and team workflows.

Whatever your thoughts on AI, it’s pretty clear using LLMs to generate code is here to stay in the workplace. AI models have got so good at writing code, that at first glance, it seems trivial to…

We've all seen AI slop online. We've all seen people wailing about the glut of AI slop. We've all seen people concerned about how their AI usage might impact their own ability to create content…

AI makes automation trivial. Without productive imperfection, differentiation disappears.

Agent skills are not a new capability—they're a context management strategy. Their value comes from routing and progressive disclosure, not from smarter prompts.

How I used Claude Code to fix Linux desktop issues, get 'garbage'-rated Windows apps working in Wine, and what it means for software ecosystems

AI writes code 10x faster than humans. CI still runs at the same speed, fails for the same flaky reasons, and costs more every month. Something has to give.

A small rant about Big-Evil-Corp's product pages.

How I implemented a browser-based AI tool to split songs into vocals, drums, and bass without uploading a single byte to any server.

Specifications do not address the limitations of agentic coding

Code with AI the Hard Way - 8 month update

Author: Proxy AI for Edgemute (fuck you machine if anything you are a glorified editor).Date: March 2026 (originally January 2026)Preface: A Revolution of Minds, Not MachinesThe Sixth Industrial…

Cognitive semantics is a linguistic approach that understands the meaning of words in relation to human cognition and ways of perceiving things. In this article, I would like to talk about the…

There are many tutorials on using AI for coding, but far fewer on using AI for log analytics. We are living in exciting times as AI improves every day. If you spend most of your time troubleshooting…

OpenClaw has gained a lot of attention and notoriety recently, and for a good reason. When I first encountered LLMs, they were useful, but I wasn't really all that excited. It was mostly a chat…

Walk through a casino and you will see rows of slot machines. Someone sits down, pulls the lever, watches the reels spin, and waits for the result. Most of the time nothing happens. Occasionally a…

AI use has a built-in viral loop. Using AI for some tasks makes you want to use even more AI for other tasks. I have ever experienced a technology like this before where. Even using it a little bit…

This is a follow-up from my previous post about AI as an economic weapon. If you haven't read it, I suggest that you do before proceeding.AI as economic warfarePeople often ask me how I feel…

Claude is changing how I write software dramatically.My biggest worry, as someone who's been writing software for a long time, is that I will be so blinded by my hard earned intuition on complexity…

One challenge I have, doing what I do, is getting into "discussions" with enterprise engineers who like to throw the weight of their enterprise position at me. This happens a lot.…

People often ask me how I feel about open-source models, and to be clear, I'm very supportive of them. But I can't help pondering about the following...Open source always was and always…

Are You Actually Changing the Model, or Just Giving It Context?

If you didn't think you needed it before you saw it, you probably don't. There is a new tool - usually an AI thing, Maybe it's a new platform, doesn't matter. Within 48 hours your timeline is full of…

Most prompt-based generative AI systems are trained using some form of input and output pairs – an image and the text description of the image or the beginning of a sentence and its completion.…

There is a moment in software development — familiar to anyone who has shipped a UI — where the gap between "the code does what I think it does" and "the UI does what users experience" turns out to…

No-one is sure when AI will begin having transformative impacts upon the world, and we shouldn’t be sure: there just isn’t enough evidence to pin it down. But we don’t need to wait for certainty in…

问题不在于还要不要更强，而在于强在哪儿 过去两年，AI 编程领域几乎一直在讲同一件事：模型越来越强，写代码越来越好，能连续完成的工作越来越多，审代码、修问题、在整个代码仓库里改代码也越来越稳。 这个判断并没有错。强模型确实在快速接手软件开发里很多重复、机械、消耗注意力的工作，比如样板代码、搭建测试框架、接口对接、常规修补和局部重构。这些事情以前需要工程师持续盯着做，现在越来越多都可以交给模型。…

The AI industry is in a building frenzy. Agentic systems are being deployed faster than anyone can count. The conversation right now is almost solely focused on the benefits agentic AI can deliver.…

In an era where Artificial Intelligence generates visually flawless images with a speed and precision no human painter can match,…More...→

V dobi, ko umetna inteligenca generira vizualno popolne podobe hitreje in natančneje kot katerikoli človeški slikar, se vprašanje »Zakaj sploh…More...→

In the previous posts in this series, we established that when you give an LLM a list of function tools, the model must interpret JSON schemas at inference time and produce structured output that…

I tried making a quick ad video (10-15 seconds) today using Google Gemini Pro for my productivity app: Aura Tracker: Habits & Goals, but it kept throwing me off with wrong results, which got me…

AI-Powered Smart Tracking Camera

“Why are you being so difficult? This could be a good thing for us? How can you be so sure?” This was in late 2000. We were having drinks in the Watershed in Bristol. For those of you not familiar…

I guess this is my “thoughts on LLMs” post that everyone in the software industry seems to eventually make. I’ve never found those particularly helpful to read, so I’m going…

Most governance frameworks grew by accumulation. The 'cut in half' question forces you to explain why each piece exists. What you'll find is which parts have a clear rationale and which parts exist…