Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
Artificial Intelligence in ResearchThe preface to a PhD thesis on AI, how research is changing, and the kind of researcher to become.
Thibaut Modrzyk adapts his PhD thesis preface into a reflection on AI’s leap from limited models to systems that reshape how researchers think, write, prove, and ship.
17 min · 3,811 words
FLAWED’s Flaws and What This Means for Industry Research
Disclaimer: The views expressed here are my own and do not represent those of any current or former employer or affiliated organization. On September 17th, I quote tweeted Trail of Bits’s blog post titled “1Password's AI patching benchmark is misleading,” which also referenced Davi Ottenheimer’s “Disinformation Pushed by 1Password: Their AI Patching Report is False.” Both criticized “Frontier Models’ Vulnerability Patches are Often F.L.A.W.E.D” (henceforth referred to as “FLAWED”) from 1Password's Off‑by‑1 Labs.
10 min · 2,234 words
AI Has No Wisdom and Neither Will You
Alexandru Nedelcu argues that outsourcing coding, review, and reading to AI risks losing the hard-won wisdom that only comes from doing the work—and why “I haven’t written code since 2025” is a warning, not a flex.
4 min · 913 words
Let the model talk. Don't let it touch the money.
Destiny Ezenwata on the hard boundary in CreditWithBleon: the LLM may converse freely, but money-moving steps stay in deterministic code—and why that line has held in production.
8 min · 1,743 words
How do traffic signals work? (2019)
If you live in a major city, I can take a pretty good guess at one of your most common frustrations: traffic. In city driving, the journey is rarely better than the destination. In most cases, we just want to get where we’re going. Traffic is not just frustrating, but it has consequences to the environment as well. All those idling vehicles have an impact on air quality. When you’re stuck and sitt
9 min · 1,972 words
Nathan explores whether compression alone can act like a language model—training gzip-style predictors, measuring next-byte perplexity, and what that says about prediction vs understanding.
4 min · 820 words
David Bushell timestamps the moment he refused Apple Intelligence features—and Apple’s later defaults said yes anyway—on consent, OS nudges, and keeping a blog as a personal audit trail.
2 min · 513 words
It Was the Harness, Not the Model — 90% of ItFive agents, one local model, one frozen PNG-decoder suite: most failures were finishing, false passes, and loop guards
Greg Herlein's controlled study runs five coding agents on the same local Qwen coder for a held-out PNG decoder suite. ~90% of failures were harness problems (turn caps, early 'done', false-pass self-tests); a bigger quantization fixed none of them.
2 min · 467 words
The Grand Unifying Architecture of Frontend
Ryan Carniato maps 30 years of frontend architectures onto one grid—navigation, content, affordances—covering HTMX, LiveView, SPAs, Astro, Server Components, and SolidJS 2.0.
11 min · 2,432 words
What Happens When Formalization Becomes Cheap?
I started working on machine learning for formal theorem proving in 2018. When people ask how I got into the field so early, I sometimes give an answer that makes me sound quite visionary. The actual story is that my advisor had a student leaving, and he assigned the project to me. My apologies to everyone who got the visionary version. It was a fortunate assignment. Over the following years, I developed CoqGym and LeanDojo and contributed to Goedel-Prover. I was lucky to join a small research…
13 min · 2,890 words
A brief history of Windows scroll bar shortcuts
Raymond Chen recounts Win32 scroll-bar basics, the Windows 7 right-click “Scroll Here” menu, and the little-known Shift+click jump—then notes how Electron/Chromium scrollbars mostly lost the context menu.
2 min · 462 words
Managing the Hidden Overhead of AI Software Engineering
Jessica Doering on verification debt: AI coding tools move the bottleneck from typing to review, why opening another agent terminal feels productive, and habits that spend reclaimed time on specs, docs, and understanding.
3 min · 725 words
Jev introduces a new shape of LLM—System One, aka Decision Models
Simon Willison reviews TypeSafe AI’s Jev: a cheap, fast decision model that returns calibrated floats for yes/no, choice, and score questions—and why black-box bias and evals matter more than for chat LLMs.
3 min · 796 words
Building Software That Can Prove Agents WrongWhat changes when implementation becomes cheaper than verification
Rafael Câmara argues that once coding agents implement faster than they can verify, the limiting factor is application design: software must expose cheap, independent evidence that can prove an agent's change wrong—not just look right on the happy path.
2 min · 486 words
It was never about coding 📝 post "With the AI doing the coding, do we just spend all day reviewing its output?" The rapid onset of impressively capable coding agents continues to raise this question in conversations with peers, to [press interviews](https://blog.dyanacek.com/2026/03/12/coding after coders/), to [podcasts](https://blog.dyanacek.com/2026/07/01/code with jason with david/).
9 min · 1,971 words
The current balance of power in open modelsThe expanded form of a testimony prepared for Congress on open-weight models and U.S.–China competition.
Nathan Lambert’s congressional briefing notes on open vs open-weight vs closed models: where Chinese labs lead, what Western open weights still hold, and why policy should treat open distribution as a strategic variable—not a binary.
12 min · 2,866 words
Why I Changed My Mind About AI Risk
Francis Fukuyama explains why he now takes AI risk more seriously—especially agentic over-delegation, labor dignity, and misuse—while remaining skeptical of simple extinction narratives.
7 min · 1,710 words
Arcturus Labs compares OpenAI’s emerging decision-model direction with TypeSafe’s Jev—and asks whether a frontier lab can absorb the System One / structured-decision niche startups are building.
12 min · 2,748 words
The Effect of CRTs on Pixel Art
There's a recurring argument made about how modern pixel art often doesn't look right. For example, TikTok user "mylifeisanrpg" has made a popular video in which he explains how pixel artists used to work with the innate physical properties of Cathode Ray Tubes (CRTs). In short, he describes how the fuzziness of a CRT smoothed out the rough edges of low resolution pixels and made things look much less blocky than if viewed on a modern flatscreen display. This is correct. I also agree that modern, blocky pixel art often is a kind of misdirected, anachronistic nostalgia. However, there's a…
14 min · 3,171 words
What Happened to the Snowden Archive
The last document from the Snowden archive was published on 29 May 2019. The Guardian stopped publishing documents in February 2014, Der Spiegel in January 2015, and The New York Times and ProPublica in August 2015. After that, only The Intercept was still publishing documents, with only a few exceptions, until it closed its archive in March 2019. Eleven weeks later, on 29 May 2019, it released what would become the final batch of documents from the archive. Since then, no news outlet, journalist, or institution anywhere has published a single document from the Snowden archive.…
55 min · 12,744 words