The adoption record

Building a system is half the job. The other half is getting people to actually use it. This page is the record of that half: rollouts, training, and teaching.

As executive assistant to the Chief Nursing Officer at National Jewish Health, I replaced paper schedules across seven staffing areas with cloud, formula-driven workbooks. Each manager got several hours of administrative work back a week, and staff got real flexibility in how shifts get filled.

The float pool is one of the seven areas, and the one I later rebuilt as a purpose-built app: no-login staff links, a written handover, and a test suite that gates every release before it ships. It is in the hospital’s IT security review now, not yet in use with hospital data.

I’ve trained colleagues at National Jewish Health on using AI tools, and I mainly use AI to advance my own workflows, on non-patient work only. No patient data goes into an AI tool at any point.

I taught Hack Reactor’s live online JavaScript program: 20+ hours of live lecture and live coding a week, and mentored 150+ students through data structures and algorithms. Teaching people a new way of working, at that scale and cadence, is the same muscle as getting a team to adopt a new tool.

The float-pool scheduler ships as staff links with no login: nobody has to remember a password to use it. Training waits until the tool clears review, and a written package answers the hospital’s security questions item by item rather than leaving them to a meeting.

The same pattern shows up in what I’ve built for others to run: Too Great to Go’s spec is written so the system could be rebuilt from nothing, and the float-pool app comes with an operator manual and ten incident playbooks.

How I work with AI covers the build side: the six-step method, the gates, and what I won't automate.