My First Week With GPT-6 Astra
GPT-6 Astra was my main Codex driver for the last week, and I am back on GPT-5.5. That sounds harsher than my actual opinion…
Read postBlog
Technical notes from real projects
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.
73k readers last year
310+ published posts
Laravel, AI, SaaS, SEO
Latest post
Older posts
Continue browsing the archive.
AI Will Not Make Everyone More Productive
When everyone gets access to AI, not everyone becomes more efficient. Some people and companies will become much faster while some may become even…
Read postThe End of Claude Code Automation
Starting June 15, Anthropic is moving automated Claude Code usage out of the normal subscription bucket. claude -p usage will now count against a…
Read postClaude Code’s Hidden Advisor Tool
In a typical multi-agent setup, the smartest model is in charge. It plans, decomposes the task, and sends smaller pieces to cheaper workers. The…
Read postWhy I Do Not Recommend Fastmail for Primary Email
Ask for an email recommendation on Hacker News or Reddit, and everyone agrees that Google spies on you and Microsoft is clunky, so the…
Read postSoftware Development Is Becoming a Factory Job
When factories automated their assembly lines, the repetitive tasks went to machines, but humans stayed to oversee, fix problems, and make decisions. That's exactly…
Read postIs Opus 4.7 a Downgrade?
Opus 4.7 is not generally a worse model than Opus 4.6, but there is a real downgrade: with Opus 4.7, the control over the…
Read postWhy No One Reads Your Upwork Proposals Anymore
For years, Upwork worked in a frustrating but understandable way. You applied to dozens of jobs, most went nowhere, but a few clients opened…
Read postGoogle Search Is Broken
I searched for my own name on Google. The result makes no sense. When I search for vincent schmalbach, Google gives me only two…
Read postOpenAI models answer an older prompt instead of the current one
There is a strange failure mode that I have been noticing in OpenAI models: in a longer session, the model can suddenly answer an…
Read postHow Codex CLI Flags Actually Work (Full-Auto, Sandbox, and Bypass)
If you use the Codex CLI, you’ve probably seen a few flags that look like they do the same thing: --full-auto, --sandbox, and --dangerously-bypass-approvals-and-sandbox.…
Read postAI in the Mining Industry
Mining is currently executing a massive pivot from manual, labor-intensive extraction to a model driven by data and automation. This is a fundamental shift…
Read postMaking Money From What You Care About
There's this idea floating around that if you're top 1% at anything, you'll make money. It sounds nice but it's not quite true. What…
Read post