Claude-Code on rothdaniel.de

Claude Gets It Wrong 67% of the Time and I Still Can't Stop Using It

Sat, 04 Apr 2026 00:00:00 +0000

I keep detailed data on my Claude Code sessions. Not by choice; Claude Code stores session metadata including friction points. I pulled the numbers from 55 tracked sessions and the results are worth writing about.

The uncomfortable numbers

Out of 55 sessions with tracked outcomes:

37 had buggy code output
37 involved Claude taking the wrong approach
7 had Claude misunderstand my request
3 involved excessive changes I didn’t ask for

That’s a 67% rate of buggy code and a 67% rate of wrong approaches. Those numbers sound like a tool you’d stop using. I haven’t. I use it more every month.

From 12 Prompts to 960 in a Month: My Claude Code Adoption Curve

Sat, 28 Mar 2026 00:00:00 +0000

I have 1,714 entries in my Claude Code history file spanning six months. The distribution is absurd:

Sep-Dec 2025: 12 entries total
January 2026: 352 entries
February 2026: 960 entries
March 2026: 391 (and counting)

I went from “trying this out” to “daily driver for everything” in about six weeks. Here’s what happened at each stage, and what finally made it click.

Stage 1: The skeptic (Sep-Dec 2025)

12 entries in four months. I’d open Claude Code, ask a question, get an answer, close it. It felt like a fancier Stack Overflow. I wasn’t giving it enough context to be useful, and I wasn’t staying in the session long enough for it to learn about my codebase.

How I Ship Side Projects in Weeks Instead of Months with Claude as My Rubber Duck

Sat, 21 Mar 2026 00:00:00 +0000

I wrote my first post about AI back in February. It was high-level; how AI fits into my workflow, advice for new engineers, that sort of thing. This post is the receipts.

I pulled every session file from my local Claude Code install. 222 sessions, 1,714 history entries, 615 of my own messages, 8,117 from Claude. Six months of data. Here’s what it looks like when you treat Claude as a pair programmer for side projects.