Switching from GPT-5.5 to GPT-5.6 Made Me Less Productive
I pay for three Codex subscriptions at $200 each, and for the past week they have mostly bought me waiting. Since I switched from…
Read postBlog
Laravel, AI-assisted development, SaaS, SEO, tooling, and the parts that only show up once software is used by real people. Mostly technical, but not trapped in one narrow lane.
The newest notes, fixes, and observations from the blog.
Give me a few hundred million dollars and a year and a half, and I will build you a pretty good LLM. Not a…
Read postTIME is now serving two different versions of its website. Humans get the magazine. AI crawlers get a stripped down markdown copy with ads…
Read postGPT-5.6 Sol xhigh now uses more than twice as many tokens per session as GPT-5.5 xhigh in my Codex workflow. For Codex users, tokens…
Read postLLMs use Markdown tables all the time. A few months ago, Codex CLI displayed them as ordinary wrapped text. In a narrow terminal, the…
Read postGoogle Lighthouse and Cloudflare are adding different controls for the same emerging class of automated web visitors. Lighthouse added an experimental Agentic Browsing category.…
Read postGoogle's Gemini 3.6 Flash and Gemini 3.5 Flash-Lite ignore temperature, top_p, and top_k, according to Google's current migration guide. Existing Gemini requests could send…
Read postStart with the area closest to what you are working on.
Browse every topic on the blog.
Recent posts grouped by the topics I write about most often.
Everything published on the blog, newest first.
I pay for three Codex subscriptions at $200 each, and for the past week they have mostly bought me waiting. Since I switched from…
Read postGive me a few hundred million dollars and a year and a half, and I will build you a pretty good LLM. Not a…
Read postTIME is now serving two different versions of its website. Humans get the magazine. AI crawlers get a stripped down markdown copy with ads…
Read postGPT-5.6 Sol xhigh now uses more than twice as many tokens per session as GPT-5.5 xhigh in my Codex workflow. For Codex users, tokens…
Read postLLMs use Markdown tables all the time. A few months ago, Codex CLI displayed them as ordinary wrapped text. In a narrow terminal, the…
Read postGoogle Lighthouse and Cloudflare are adding different controls for the same emerging class of automated web visitors. Lighthouse added an experimental Agentic Browsing category.…
Read postGoogle's Gemini 3.6 Flash and Gemini 3.5 Flash-Lite ignore temperature, top_p, and top_k, according to Google's current migration guide. Existing Gemini requests could send…
Read postI tried to make AI writing sound more human with logit_bias, an API setting that changes how likely a model is to select specific…
Read postA research agent can leak a private document without uploading it. It can read a detail, turn it into a web search, and send…
Read postOn July 22, 2026, Qualys disclosed CVE-2026-64600, a race condition in XFS, a Linux filesystem. The flaw is in XFS's copy-on-write path, where the…
Read postGoogle’s Managed Agents, now in Public Preview, can give a Gemini agent a Google-hosted Linux sandbox. Google documents the default Antigravity agent as able…
Read postOn July 20, 2026, OpenAI described an incident in which an unnamed internal model found many stuck find processes and every shell command hanging.…
Read post