2 posts tagged with "ai agents"

For at least the last year, I haven't written implementation code by hand; I usually first see it in pull requests. This post covers how I work with coding agents today and why context, taste, prioritization, and reviewable scope matter more than ever.

Following up on my 2025 workflows, this post explores the shift from manual prompting to agentic feedback loops. I break down why model choice now outweighs prompt engineering, how Claude Code has become my primary driver alongside Cursor, and why the "SWE II" level of AI requires a new focus on automated eval loops and organizational standards.