A growing library of one-day AI projects — with the time, tools and results for each.
Claude Code vs v0 vs Lovable: I gave three AI tools the same app brief. One I built and measured, two I judged on documented behavior. Here's the winner.
I ran one workflow through n8n, Zapier, and Make. The winner isn't about features, it's about how each one bills you. The 2026 math.
AI vs human web development: I hand-built a one-page site, measured a real client tweak, and set it against what AI builders are documented to do.
Read by category
NotebookLM Data Tables promises to turn messy data into clean tables. I tried it with a real messy spreadsheet. The catch is at the input, not the AI.
Automating your work with AI starts with one boring task, not your whole job. How to pick it, choose a tool, and set it up in an afternoon.
I gave five AI writing tools one brief: write my bio. Each broke differently. Where ChatGPT, Jasper, Copy.ai, Grammarly and Sudowrite fall short.
An honest Notion AI review: where it removes work, where it just moves work to the editing seat, and why the real price is $20 per user, not $10.
A Sora 2 test review that skips the hype: a verified map of where its physics promise breaks (hands, flips, text) and why the app closed in 2026.
I ran a day of errands with an AI browser agent. It nailed the finding half. The buying half is exactly where you keep your hands on the wheel.
Claude Cowork vs Claude Chat: Cowork takes real actions on your files. What it does that chat can't, what it costs, and who should skip it.
I tried to set up a freelance CRM with AI in a day. The dedicated tool wasn't the answer, the $0 spreadsheet nearly was, and here's where it broke.
I checked 5 Canva alternatives, and 'free' means three different things. Two need no account. One is really a trial. Here's the honest sort.
Set up your first AI agent in one afternoon, free and no code. Plus the three places a beginner's agent always breaks, and how to dodge them.
Gumloop vs n8n: I counted the AI nodes in both public registries. n8n has 122, Gumloop 15. The agent gap isn't real. The license line is.
I measured 15 production Claude Code skills, file by file. 81% of the library stays out of the context window until it fires.
That's the latest. Explore the full library: