I don’t think we are ready to NOT read the output I read the code for things that are important. Personal projects, non-critical, funsies - sure I’ll allow it through and take the output…

Two of the world’s most influential regulatory bodies picked the same day — August 2, 2026 — to tell you the same thing: you have a right to know when the thing in front of you came out of a…

mcp-data-platform exposes knowledge as wiki-style pages. An agent talking to it over MCP can search them and fetch them, and the pages can point at each other the way an internal wiki does. The…

New AI difficulty: Brutal, between Hard and Deity The AI improvements in this release made every difficulty play sharper, so the old Hard tuning now lives at Brutal: bonus yields, a free starting…

Accessibility helps people use the web directly and gives browser agents a better foundation for assisting them - without taking control away.

Using an LLM agent to bring productivity metrics into engineering retrospectives

Disclaimer: Prompt injection is an unsolved problem.

Two years ago I was a skeptic about GenAI. Now Emacs has become my agent cockpit.

I started Transformer Lab because I wanted to dedicate my career to the work that could have the largest possible impact on the world. I would like to take some time to walk through why I believe…

I work on security detection at Datadog. The model I have been building, Mambark, reads audit logs the way a language model reads text and scores every event by how surprising it is. It is a small…

ArchSpec turns your architecture into an executable spec: declare components and boundaries in one Ruby file, and every change gets checked, whether a person or an agent wrote it.

I miss being an SRE in the Old Days. Here's how we do release management with agents now

A three-week test of OpenAI's ChatGPT ads beta for a B2B WordPress product: how the platform works, what it costs, the results we measured, and who it's actually for.

As AI makes database operations easier, the enduring role of database tools may shift from UI to governance.

RollTab is a real-time piano MIDI autocomplete model and iPhone app.

# Is your model smarter than a 5th grader? I saw a post about [LittleLearner](https://littlelearner-ll.github.io/), a scale-up study where Qwen3 dense models are pretrained on a corpus filtered down…

There’s little question that SUNO, and products like it, will change the way music is produced at scale. Sure, we can have the discussion about what’s best for music, for listeners, for…

With the increasing availability of assistive AI, many people have turned to AI frameworks and harnesses (Hermes Agent, Claude Cowork, etc.) to help them achieve day-to-day goals. In many cases the…

Well, I’ve gone from initial, curmudgeonly reluctance, to shruggy acceptance, to full-blown, guns-blazing, AI mania at work. I’m writing skills, shipping agents, thinking through internal…

With every day that passes, another model joins the ranks at the frontier. Last week it was xAI's Grok 4.6. This week it's GLM 5.3. A month ago it was Kimi K3. And it clearly won't be…

After a recent conversation where someone mentioned that they experienced multiple model lineages surfacing additional issues in code reviews, I decided to revisit the code review machinery in my…

Sometimes it will. But often you upload a song you own asking Suno to extend it or flesh it out, and you get the “uploaded audio matches existing work” rejection. It’s not a bug.…

The AI industry is like Dracula, sucking blood from our society. You'll never guess who's coming with the stake.

I wanted everything Claude Code writes for developers to follow the Google developer documentation style guide. A forked skill, a few adaptations, and a local snapshot of all 70 guide pages got me…

Some things take real talent to be that bad — you laugh, you mock, you almost envy, until badness stops being bad and becomes the absurdity of life itself.

Twenty minutes after I told a coding agent to use GLiNER to structure a set of messy records, it said it was done. That was too fast. I opened the code. It had parsed the raw fields with regex and…

We are seeing a Cambrian explosion of open-source software applications.I have written before about how AI has made software dramatically easier to build. A developer can now create in days what…

AI is the new "organic" in the job market. Every resume claims to be an expert in all things AI. The valuable person is not listing AI as a signal. It is the person who combines AI skills with…

A friend posted “AI has no place in the creative space. Period.” That isn’t the most ridiculous thing I’ve ever heard, but it’s close. Excluding AI from creative…

This post is also available as a video on the Modern Software Engineering channel Kent Beck once said – ‘I’m not a great developer, I’m a good developer with great habits’. Now, Kent Beck…

What the Agent attempted with DNS in the Hugging Face–OpenAI Intrusion Incident of July 2026

I started working in technology around 1993. One of my first jobs was in quality assurance, partly because there was no security profession to join yet. There were people doing the work, but few…

This is a short story of how thanks to AI agents, I am again passionate about my open source projects, namely pretty-git-prompt (p-g-p in short).

Twenty years ago I participated in NaNoWriMo. I was just remembering how I had a printed book that was all synonyms and antonyms. Handy when I got stumped on just the right word to use. That’s…

Remember when I told you all I'd swim in one lane? Well... I'm swimming in two to six lanes now. And I can report that my deploy paralysis is gone, and that my mental fatigue doesn't kick in meaningfu

how an 18-month standards effort quietly turned into the compliance layer for AI-to-CRM access

Last year I laid out four pillars driving the longevity boom. January started strong validating three of them. Eight months later, all four have moved forward, but they’ve hit the same wall we’re…

Software rarely fails because it lacks a feature; it fails because it cannot scale, cannot recover, or cannot be securely maintained. Yet, Non-Functional Requirements (NFRs) are routinely treated as…

Please welcome back Lila Warren, a recent contributor to the 21st Century Tech Blog. This is her fourth article appearing here. All of them have been related to artificial intelligence (AI) and its…

When an agent writes the code and the tests, passing doesn't tell you much. Black box acceptance tests are the independent verification boundary that actually matters.

Having been robot-wrangling for a solid year now, I continue to feel like they give me superpowers. However, like all of us, I have a front-row seat to increasingly attractive garbage the robots are…

T3 Code provides a convenient interface for working with coding agents such as Claude Code and Codex. One of its more interesting features is that the T3 client doesn’t need to run on the same…

Every serious political and economic argument about AI eventually collapses into one question. Not left vs. right. Not West vs. East. Not safety vs. acceleration. The real divide is simpler: does…

There is so much fluff out there telling us that we are nearing the limits of our current technology. I would argue that couldn’t be further from the truth. As...

Simon Goldstein has just published the first paper in a top philosophy journal that was openly written primarily by AI: “Epistocracy and the Commitment Problem,” published in Philosophy…

In the past, new PhD students would ask how they could succeed when they had to compete with the likes of say, Richard Karp or Avi Wigderson. I would say Karp and Wigderson have limited bandwidth and…

AI agents can code, dig through customer data, and fire off tools faster than any person. The problem is they also get fooled. A prompt injection or a random glitch can send one of them racing to…

I resigned. My last company, which worked in regulated industries (e.g. Defense) projects, became AI-first late December. Regular readers know my tiredness for genAI, its ethical failings, and how I…

automating job applications has almost killed the usefulness of applying. quality over quantity.