Tag: ai
Pook-Emu Bee: Links For 07-14-26
Two consecutive editions of Pook-Emu Bee daily links? What in the world is going on here? [caption id="attachment_6415" align="aligncenter" width="1275"] Seen here is the Brooklyn Bridge Park Marina on July 13, 2026. I lined the American Flags, so I took a couple of photos. To the left you see Manhattan and off in the distance...
Own your AI with local models and open protocols
A Local-First Conf talk about local models, open protocols, and user-owned AI.
The shadow copy was the wrong layer
Last week I set out to build two companion WordPress plugins. One would turn PDF attachments into markdown so AI agents could read them. The other would run a local vision model over every image and write a description an agent could use. Generally speaking, agents can't read PDFs or images well (although they're getting...
Kreativität kennt keine Abkürzungen
Bei der Teilnahme an einem Kreativworkshop wird man oft darauf hingewiesen, dass ein Produkt, das scheinbar Kreativität verkörpert, in Wirklichkeit das Ergebnis harter Arbeit ist.
ATProto MCP server is now functional
The ATProto MCP server was wired into a working state with public and authenticated Bluesky tools, tests, and docs.
Remote Development from Any Browser with blit.sh and moo
How to pair blit's browser-based terminal streaming with moo's local-first agent workbench for remote development, 11ty site work, and accessible document and transcript conversion.
Our session replays are opening their own pull requests
Errors are the easiest data to act on in our product improvement pipeline . They arrive with a stack trace that's groupable and says exactly what broke and where. This makes it easy for agents to research each one, and if it's actionable, open a PR…
Knowing which Claude is stuck
A zellij session switcher that shows how many Claude Code agents are running in each session, and which of them has stopped and is waiting on me.
Code For the People
New short documentary (< 20 minutes) via Automattic about Open Source and the Future of the Web. If you're already working in open source there won't be anything new here, and if you're part of the WordPress community you might even wonder if the timing of this has anything to do with the WP Engine...
Weeknotes no.3 – how AI-pilled are you? 🧠💊
Taken the ai fluency index test?
Open Source Project: Why I Built the ScribeCMS ecosystem on AT Protocol
The story behind Scribe: Why I rebuilt my CMS on AT Protocol, the stack decisions that went into it, two moments that validated the approach, and what's next on the roadmap.
Why Human Thinking Still Matters
We're building tools to save time, yet we often find ourselves with less time to think about why we're building them. A reflection on reclaiming the "messy" human part of the development process in the age of AI.
Governance can prevent AI from being used to undermine democracy. But only if it has teeth.
The apparatus to cement centralized, undemocratic power has already been built. Regulations have a part to play, but they can't just be recommendations.
A Case for the Ethical Treatment of Generative AI
Are there ethical implications of generative AI consciousness? From von Neumann's self-replicating machines to modern LLMs, we must inventory the problem of other minds, the possibility of AI consciousness, and why we ought to consider the welfare of systems that may experience something like suffering.
"You have to put your body into your writing".
You have to put your body into your writing. A long time ago before I came to tech, I wanted to work in the creative industry and I was assistant to an editor. He was toxic in many ways and I ended up running away from him, but that advice from him still stays with me.He said he could immediately tell when somebody copy-pasted text and stitched it together because he didn’t see themselves inside the writing. Obviously you cannot put your body into writing, but you probably understand what it mea...
From Bioreactors to Brier Scores: Building Model90, a Football Prediction Engine
What happens when a bioreactor engineer points his state-estimation toolkit at football. Model90 is a statistical forecasting engine for the 2026 World Cup and the major leagues, built on Dixon-Coles Poisson, Elo, xG and a calibrated meta-model. An honest look at how it works and what its Brier score really means.
How Kamal Fits My Coding-Agent Workflow
Why I run small products on one Hetzner server with Kamal, and how Agent Control serializes deployments when coding agents are doing the work.
Chronicler v1.8.x — LLM newspaper generator and issue backfill
Chronicler shipped v1.8.2 with a configurable LLM newspaper generator, server-log issue backfill, and an AGPL license.
Weeknotes no.2 – hello AT Proto & local LLM 🦋🦖
Added AT Proto publishing to my blog and played with local LLM models
The Programmer’s Fulcrum: 3 July, 2026
This post originally appeared on The Fulcrum. Welcome to this week's The Programmer's Fulcrum. It's your weekly curation of the essential news in the Open Media Network and open social development communities with a focus on devastating big tech via Techno Anarchism. As usual, we aim to provide actionable content you can use to destroy...
Why I Built an Orchestration Tool for My Coding Agents
Why I built Agent Control, my macOS app and CLI for coordinating coding-agent sessions, queues, callbacks, loops, and cross-project work.
Settings in “No AI” DuckDuckGo
Back on June 8, 2026, Chris Wiegman published a blog post titled You Don't Need the DuckDuckGo noai Subdomain. For those not in the know, noai.duckduckgo.com is a subdomain of DuckDuckGo search which removes DuckDuckGo's new "AI" features. I wrote about how to add the "noai" version of DuckDuckGo as a search engine to Chromium-...
Making the Impossible Possible
I find myself waiting on the blinking square cursor of the Claude Code TUI thinking "go faster," even though I know it's already going so much faster than I would have been able to go myself a few months ago. Since starting development with AI agents I feel there is so much more to do...
OpenAI wants to give us 5% of its success. It's a bad bargain.
A wealth fund that shares 5% of AI success with government and voters is either based on hype or not nearly enough to cover the damage. Either way, the incentives are perverse.
Draw me like one of your JSON files
In Pixel Pushing, Redux I rebuilt this site with Claude Design over a weekend and £23.84, then handed off to Claude Code for the WordPress build. One thing intrigued me when setting up the design, Claude Design's first pass was to set up a design system, you can import assets and it will set up...
I have a theory about AI fake news site The Editorial
I was taken in by a fake AI news site. I think it may be an attempt to poison LLMs with propaganda.
Pook-Emu Bee: Links For 07-01-26
I have not published "daily" Pook-Emu Bee links since June 23. But today is the start of a new month. Let us turn over a new leaf in July with the Pook-Emu Bee. [caption id="attachment_6173" align="aligncenter" width="1008"] I took this photograph in Brooklyn Bridge Park on the morning of July 1, 2026. It is difficult...
AI's costs are going through the roof - so businesses are telling LLMs to talk like cavemen
Me think this hilarious sign of times.
I don't know how to evaluate coding models
The rate of LLM improvements for coding keeps increasing, and I can't keep up.
letta-writer
Memory repository for a Letta Code writing and document agent.
Cal Newport on Big AI “Doom Trolling”
Cal Newport wrote a New York Times op-ed calling on big AI companies to stop "doom trolling" and published a short summary of his argument on his personal website. He specifically criticizes some of the big AI companies for warning (or threatening?) that their technology could spiral out of control and put millions out of...