Residency education and assessment
The case they are scrubbing into tomorrow becomes the board question they get right next year.
Faculty name the case. ResidentReady retrieves the sources, writes the questions against them, verifies each one independently, and texts the resident a short quiz. What comes back is a study pack and a measurement that accumulates.
1
Faculty enters the case
An operation, a diagnosis, a consult, a topic. Pick residents, level and length. About thirty seconds.
2
The pipeline builds the quiz
Sources are retrieved, questions are written against them, then an independent pass tries to break each one. Items that survive are released.
3
The resident gets a text
Tap the link, answer five to ten questions on a phone. No password, no app.
4
A study pack lands immediately
What they missed, the core concepts, board and case pearls, and the sources behind each claim.
5
It compounds
Every answer feeds domain-level performance, retention checks and a longitudinal view for the resident and the program.
What is deliberately not claimed
Most of the value in an assessment product comes from what it refuses to say.
- No question ships from model memory. Every scored item is written against retrieved, allowlisted sources and then attacked by a second pass. If the evidence is thin, you get fewer questions — or a refusal, and an explanation of why.
- No board prediction. The longitudinal figure is percent correct on ResidentReady board-domain questions, named as such, and it does not render at all below a minimum sample.
- No invented percentiles. A comparison appears only when the group is large enough, and the group is always named.
- No patient data. Topics only. The pipeline refuses input that looks identifiable.
Specialty-agnostic
Twenty-one residency specialties are configured, each with its own training length, in-service and board targets, knowledge domains and source policy. The triggering object can be an operation, a diagnosis, a consult, an imaging or pathology topic, or a board domain.
Formative, not evaluative