01
Teacher authority
Teachers review, edit, replace, and correct plans, assessments, and interpretations.
Principles
The rules the product is built around.
Learning intelligence touches teachers, children, and families. That demands more than good intentions—it demands explicit, inspectable commitments.
Responsible by design
Learning intelligence without turning people into metrics.
Cuegence is designed to improve academic decisions while protecting teacher judgment, learner dignity, and the boundaries between school communities.
02
Pseudonymous by design
Learning intelligence uses school-scoped identifiers rather than direct student PII.
03
No ranking
No student rank, peer comparison, teacher scoring, or automated personnel judgment.
04
Auditable intelligence
Evidence, provenance, confidence, and model revisions remain inspectable.
05
School-level isolation
One school’s operational data never informs or benchmarks another school.
06
Model-agnostic
Foundation models are replaceable infrastructure, not the durable product core.
Direct PII stays out of model calls.
Identity and contact information remain in school systems wherever practical, separate from the learning-intelligence layer.
Hard product rules
Commitments the product is built around.
These are not aspirations—they are design constraints every Cuegence feature must satisfy before it ships.
The model is the product core
Every product surface should improve the longitudinal learner model or use it to make a better academic decision.
Teachers remain the authority
Classroom teaching, lesson plans, assessment, and interventions stay under teacher control.
≤5 minutes per class per day
Routine Cuegence-specific teacher administration targets five minutes or less per class each day.
No student-facing product in v1
There is no direct Cuegence student app, AI tutor, or child-facing chatbot in v1.
Supportive parent guidance only
Parents receive practical learning support—never hidden model judgments or performance surveillance.
No ranking or gamified pressure
No student ranking, peer comparison, or pressure-inducing performance gamification.
Explainable and auditable
Recommendations and learner inferences must be explainable to authorized teachers; inferences never overwrite source evidence.
Facts about teaching, not judgments of teachers
Cuegence reports factual teacher and class information; it does not judge or rank teacher quality.
Strict school isolation
One school’s operational data never informs or benchmarks another school. No cross-organization learner portability.
No direct PII to models
Learner intelligence operates on school-scoped pseudonymous identifiers; direct PII must not be sent to LLMs or models.
The product test: a Cuegence feature should improve the longitudinal learning model with useful evidence, or use it to make a better academic decision. Commodity AI features that do neither don’t make the cut.
Privacy & identity boundary
Learning intelligence without student identity.
Cuegence does not require direct student PII for its learning-intelligence functions. Identity and intelligence are kept deliberately apart.
Pseudonymous learner keys
A school-scoped roll number or equivalent pseudonymous identifier is the learner key visible to the intelligence system—not names.
Identity stays with the school
Names, parent phone numbers, email addresses, and other contact data remain in school systems wherever practical.
Minimized model inputs
LLM and model calls use minimized academic signals and pseudonymous identifiers. Direct PII must not be sent to models.
Still treated as sensitive
Pseudonymous learning records remain sensitive personal data, with formal privacy, security, and retention controls.
Scope, stated plainly
What Cuegence is not.
Clear boundaries make a product trustworthy. These are the things Cuegence deliberately is not—and will not quietly become.
Not an ERP, LMS, or fee system
Cuegence is not a school ERP, learning-management system, attendance system, or fee-management product.
Not a generic AI copilot
It is not a generic AI teaching copilot or lesson-plan generator working from curriculum prompts alone.
Not a replacement for people
It does not replace teachers, academic coordinators, or school leadership—it informs their decisions.
Not a student chatbot
There is no student-facing AI tutor or general-purpose child chatbot.
Not teacher surveillance
It is not a teacher scoring, ranking, surveillance, or automated performance-management system.
Not a benchmarking product
It does not rank students or benchmark schools against each other—and never will.
Product north star
Continuously improve the quality of academic decisions by maintaining an evidence-backed, longitudinal understanding of learning.
That model should make teacher preparation, assessment, intervention, personalized home practice, and school intelligence progressively more specific to the learners actually being taught—without replacing teacher judgment, exposing children to performance labels, or turning teachers into metrics.
Pilot programme
Evaluate the boundaries for yourself.
Pilot schools can inspect how evidence, inference, and identity are handled end to end.