The procurement questions nobody asks about AI in teaching
Eleven questions to put to any vendor, including us, and what a good answer sounds like. Written for the person who has to take this to a committee and defend it.
Read the guide →Why a mark tells a tutor almost nothing
The gap between assessment data and teachable information, and what has to be captured instead. Includes the retrieval-practice evidence this product is built on.
Talking to your board about AI in teaching
The four objections that come up every time — jobs, cheating, data, and cost — and how to answer them without overclaiming.
Running a one-term pilot that actually proves something
Which department to pick, what to measure, and the two mistakes that make a pilot inconclusive no matter how well the software works.
What a first cohort looks like on the mastery view
An anonymised walkthrough of one chapter across thirty learners, and the three conversations it produced in week two.
Safeguarding when the tutor is software
The limits we build into Class for under-18s, why they cannot be switched off by a teacher, and what a DSL sees.
Sitting SentinelXR alongside an existing ERP
The Frappe architecture, what syncs with an SIS, and why replacing your student record system is the wrong first move.
Nothing here yet
We would rather publish nothing than publish filler. Ask Orin what you were looking for and it will point you at a person.