Skillify: Conversations Are Prototypes, Skills Are Artifacts
I manage engineering teams, but my day-to-day work is not writing production code. Most of my time is spent on the operational work that sits between engineering, planning, and execution: gathering information from multiple systems, understanding where projects are getting stuck, identifying work that needs attention, and assembling that information into something useful for me and the team.
This year I challenged myself to adopt AI tools into my daily workflow to tackle these manual tasks. Having a conversation with tools like Claude Code was particularly effective because it allows me to start with an outcome rather than an implementation. I can describe what I want, iterate on the results, refine queries, handle edge cases as they appear, and gradually converge on something that solves the problem. The process is often faster than building a dedicated tool, especially when the workflow is still evolving.