Page 4 of 14: older articles, tutorials, and videos.
The best engineers I know all have one thing in common: they know a little about a lot of things.

I explain how to automate 80–90% of business work by shifting from role-based thinking to workflow-based thinking and breaking each role into granular, automatable tasks. I walk through video editing, social media…
You still need to care about what you're building.

AI can write code, but building good software still requires ten fundamentals, and I break down the ones that matter now that syntax is automated: databases, data structures and algorithms, system design, API design…

I turn a 30-minute raw recording into a captioned, webcam-masked, music-backed YouTube upload and Twitter share in under 90 seconds with a command-line AI pipeline. I run commands to process audio, trim silences…

I automate my entire long-form content workflow with Phantom Editor, a set of deterministic Python and Bash scripts that handle audio cleanup, silence removal, webcam overlays, captions, music, YouTube uploads, and…

The programming industry has become a slop factory, and I'm tired of pretending otherwise. I point to Spotify's VP of engineering bragging about 4,500 production deployments per day when a heavy user like me notices…

I show how to caption short-form videos for free using a local Faster Whisper model to generate subtitles and FFmpeg to burn styled captions directly into the final video. I walk through the simple pipeline — raw video…

LLMs and AI coding agents still can’t automate skilled engineering work, and I use the Ghostty creator’s renderer-optimization experiment to show why human expertise remains the bottleneck. His agent ran for four hours…

Apple isn't out of the AI race just because it isn't leading the frontier—I argue that waiting while OpenAI, Google, and others burn trillions on experimentation could leave Apple in a stronger long-term position. I…

I argue that LLMs and AI agents are driving us toward cognitive surrender—a state where we stop thinking critically because the machine makes decisions for us—and that this is already visible in software development. I…
![Uber burned $3.4 billion in four months [28 million dollars a day] thumbnail](/_next/image?url=https%3A%2F%2Fimg.youtube.com%2Fvi%2FAkQpcVuP_6k%2Fhqdefault.jpg&w=3840&q=75)
Uber burned $3.4 billion on AI in four months—$28 million a day—because it set token usage as the metric and gave employees unrestrained limits. I argue that nobody needs that many LLM tokens to ship useful app…


Dude, you’re at the bottom of my landing page.
Just book a call already if you’re that interested.
You scrolled all the way here.