Ars Technica reports that an article touting the effectiveness of ChatGPT in education has been retracted by the original publisher due to questionable methodology. That analysis supposedly showed…

The fear of AI replacing your work is real. So is the shift. The teams that build structure around agents instead of resisting or rushing find the job gets better, not smaller.

If you liked this piece, please subscribe to my premium newsletter. It’s $70 a year, or $7 a month, and in return you get a weekly newsletter that’s usually anywhere from 5,000 to…

There is a scene in Jaws where the shark finally shows itself, and Brody famously says, “You’re gonna need a bigger boat.” He was wrong. They didn’t need a bigger boat. They…

Anthropic shipped managed memory stores in April. They sit at the third of four layers. The fourth, hooks, is the one that closes the Lucy loop.

(and Dealing With LLM Religion) I am opening an LLM coding contest, and I want you to enter it. Four days ago NetHack 5.0 was released, the newest major version of a 46-year-old open-source roguelike…

AI is inflicting a proliferation of tools, in the utensils’ sense. Folks are conjuring all sorts of programs out of thin air, unrestrained. Someone confessed on Hacker News to having created a…

Asimov's Three Laws of Robotics were designed as universal constraints for any thinking machine powerful enough to harm us: A robot may not injure a human being or, through inaction, allow a human…

How several Claudes and Simon Willison's example got me a Starlark interpreter in pure Python.

TL;DR: Training chatbots to sound warmer makes them significantly less accurate, whilst journalism schools lack a consistent AI strategy as scattershot policies confuse students. Meanwhile, Google,…

Building an agent sounds straightforward until you actually start. Before you write a single line of business logic, you're already deep in infrastructure decisions: How do you manage context across…

Claude Code is a terminal-based AI assistant that reads your files, runs commands, edits code, and iterates on errors. Describe what you want in plain English, and Claude Code determines which files…

I generated a landing page, looked at it, and felt nothing. Not because it was bad – it was fine. But fine is the problem. Here I share what that hollow feeling is actually telling us, why designing…

Like a young child coming home from kindergarten with their latest crayon scrawls, the internet is currently awash with people sharing their AI-generated work. And just like the young child’s…

A week between a YouTube shoot, a hypnosis session, and thoughts on impact-driven goals in bug bounty and learning in the LLM era.

Resilient Cyber Prep: Karl McGuinness The host’s framing is already aligned with your thesis: we built AuthN/AuthZ for human-paced execution; agents remove presence, pacing, and natural…

Why AI changed what I want to do after work and my ways of working

Big brain wants to create perfect setup straight away. Smol brain knows: start simple, improve later.

Nilay Patel: [E]veryone in tech understands how much regular people dislike AI. What I think they’re missing is why. They think this is a marketing problem. [S]oftware brain [is] see[ing] the whole…

I have been using agentic coding nearly every day in OpenCode. So far the MiniMax M2.5 model seems best for my purposes, as it is currently free and also powerful enough to not be frustrating (most…

Hermes Agent ships with a Kanban-style board and the Hermes Gateway that can saturate your self-hosted LLM if too many tasks are dispatched at once.

Lately there has been a lot of discussion about whether LLMs are conscious. Richard Dawkins recently added fuel to the debate by posting an article about his long conversation with Claude, which he…

If you’ve been building anything with agents in the past year, you already know the shape of the problem even if you haven’t named it: you’ve got a model in one cloud, a vector…

🔍 I Didn’t Plan to Switch I didn’t set out to replace Google. It just sort of happened. I was already using Ecosia out of habit, and then Perplexity started creeping into more and more of my everyday…

I was talking with a colleague the other day about the current industry trend to focus so much on further quantizing today's LLM models. I'm sure the reasons for pursuing quantization are apparent to…

I have talked about Vibe coding in a lot of my posts about AI and I just realized that some of the readers of my Blog Posts might not actually know what it means. ACM (Association for Computing…

How I use AI as a sounding board for home organization, while keeping every decision mine. I work in internal developer relations at Google, which means I spend most of my time thinking about how…

AbstractThis essay addresses three connected questions. First, can genius be separated from moral character? Second, does Martin Heidegger's moral failure stand outside his philosophy, or reveal…

If humans and advanced AI systems are going to cooperate—to make honest deals and avoid negative-sum conflict—AIs will need reasons to trust us. By default, they won't have many: humans routinely lie…

2026 is the age of the software factory. Dark factories where agents run the line at 3am with no humans in the building. Light factories where humans and agents work side by side at the bench. Every…

Evolutionary biologist Richard Dawkins recently wrote an article titled “When Dawkins met Claude”, where he describes his experience after two days of intense conversations with the artificial…

I recently watched a video that put a name on something I'd been feeling. The author splits an LLM's context window into two zones. There's the smart zone, where the model is sharp, and…

AI suddenly passed the “more time saved than spent” point around December 2025. A little late, I’ve finally started using agentic AI in various places over the last 2-3 months, and…

Ask vanilla RAG over Duval, Goeckner, Klivans, and Martin's 2015 paper "A non-partitionable Cohen-Macaulay simplicial complex" this question: What did Stanley conjecture in 1979 about…

Walking through spek, a small LLM-powered coding agent that turns a markdown spec into a working, tested Python package — and the six design questions I had to answer to build it.

The kernel of the AI platform I’ve been building is one Python file. Roughly 1500 lines. Tagged once at v0.6.0. Never updated, by spec. That’s a controversial design choice. Engineering instinct says…

Most AI infrastructure picks one of three: SaaS-central (one vendor’s database, one row per customer, one dashboard, infinite tenancy risk), self-hosted local (you keep everything, you also keep the…

AI Safety is theatre The AI Safety and Alignment communities are prolific. They are well funded and produce enormous volumes of research, evaluation methodology, governance frameworks, fellowship…

Open weights models keep frontier labs honest on price. If they disappear, we end up with a handful of oligopolists extracting consumer surplus.

There are ways to do voice AI without being traumatized by WebRTC.

In which he hurts the machine’s feelings First, I want to link to a few great write-ups. Your Container is not Your Sandbox - Excellent summarization of MicroVM vs containers Agentic Coding is a Trap…

The boring stuff is the point

This famous XKCD keeps popping into my mind as I watch Claude Code insisting on running stuff like find . -name "*.py" | xargs grep "function_name" across a fifty-thousand-line…

Consolidation, knowledge bases, and trying out the new wave of harness tools, a personal retrospective on April 2026.

Context Engineering for Agents banner

Companies asking 'how can we improve with AI' are really asking 'how can we improve.' The vendors selling context-ingestion are answering the wrong question because it's the easiest one to sell.

Back in the day when we wrote actual code instead of poking at an AI, I had a general rule for when to refactor repeated code. Do it once, fine. Do it a second time, fine. Do it a third time -…

Read the words as critically as you read the numbers. Exploring EU health spending data with Claude Code, the model framed government healthcare as money coming "out of people's pockets" — a slightly…