From the practitioner's field notes.
Field notes on AI and curriculum work
We Tested Three AI Models on Curriculum QA. Here's What Broke.
Same prompt, same two documents, three models, 31 real errors. No model caught them all, and two of them drifted on the second run.
Read & apply →Start with the verb, not the tool
Before you open a chat window, decide what the learner should be able to do. The tool comes second.
What a rubric is for
A rubric is a description of quality, not a scoring machine. AI can help you write one, if you tell it the difference.
Notes on alignment audits
Standards alignment is less about coverage and more about depth. A few lessons from running audits with a model as a second reader.
Differentiation without the extra binder
Three versions of every lesson is not sustainable. A single strong task with planned entry points usually is.
Keeping practice field notes
Why we write down what worked, what failed, and what the model got wrong.
New tools and field notes, in your inbox.
One new tool to try, one field note worth reading, and one idea you can bring to your team. Sent twice a month.