Skip to content
Maestro.
MaestroMaestro Overview — contents

The Concept

Adaptivity & Routing

What actually drives the adaptivity — a small set of clear, predictable mechanisms rather than something open-ended. The error type, not a score, decides the next step, and the learner participates in the choice.

The 30-second version

The adaptivity is built as a few clear flows, starting with the obvious cases and adding more as real use shows what's needed. The Evidence Check supplies a diagnosis; the adaptive engine applies it alongside prerequisites and history, and presents the learner a recommendation plus a menu of sensible options. There are eight distinct adaptive mechanisms for release one — this topic walks each.

Participatory adaptivity & agency

The governing frame is transparent, by-choice adaptivity: detect gap → explain why → offer options → learner chooses. The system surfaces a recommended next step, but also an available menu, and the learner's selection is captured as a first-class agency signal — because the platform is architected on participatory adaptivity, the learner's choices are evidence of their engagement, not noise. There is a safeguard: the recommended option is always clearly marked, so agency never becomes abandonment.

Diagnostic-band routing

Every node ends with an Evidence Check that produces a diagnostic band. The routing decision is not pass/fail — each band triggers a different next step:

DiagnosisNext stepWhy
Secure + extended-abstractOptional challenge (extend)Ready to transfer.
Secure masteryProceed to next nodeEvidence sufficient.
SlipLight feedback, then proceedUnderstanding is there.
FragileReinforce + re-checkNeeds consolidation.
Knowledge gapRemediate; route to prerequisiteConcept must be (re)built.
MisconceptionTargeted remediation for that modelA generic re-teach won't dislodge it.

Misconception-targeted remediation

When a learner's response signals a governed misconception rather than a generic error, the remediation is specific to that misconception: surface the wrong model, confront it, offer the corrected model, confirm with practice. This is the payoff of the whole evidence model — the system knows which wrong model the learner is holding, so it serves content that confronts that model rather than repeating the lesson. That is the difference between real adaptivity and a retry button.

Prerequisite & bridge routing

When a knowledge gap surfaces, the learner is routed back to the prerequisite node, or forward through a bridge node, rather than forced to retry the current node blindly. The learning graph's prerequisite, bridge, and threshold edges (authored in Studio) are what make this possible — the runtime traverses them to find the right destination.

For engineers — the three-source remediation resolver

Remediation content can come from three sources: the node's own remediation objects (authored per misconception), a routed prerequisite node, or the Companion composing targeted support at runtime. The routing engine resolves which to use based on the diagnosis, availability, and history — and always surfaces the recommendation plus the menu to the learner.

Milestone readiness gating

Each summative assessment is visible from the start, but submission unlocks only when the learner meets the readiness conditions: required nodes completed, key Evidence Checks secure, critical misconceptions cleared, learner acknowledgements complete, checklist done. The gate controls submission, not visibility — the learner can always see what they're working toward.

Misconception blockers on submission

Certain misconceptions — specified per assessment — actively block submission until cleared, so a learner cannot submit while still holding a wrong model that would corrupt the artifact. This handles a class of integrity failure that no proctoring could catch: work that superficially passes but rests on a wrong model. (More in Academic Integrity.)

Enrichment paths & Mastery Credit

When the diagnosis is secure or extended-abstract mastery, the learner is offered optional enrichment — a transfer challenge, a complexity challenge, a comparative challenge, or a contribution challenge. It never blocks progress and is framed as opportunity rather than extra homework, and it can convert into Mastery Credit — an optional academic-opportunity currency (distinct from grades) that unlocks things like SME time, showcase opportunities, or publication review.

Personalised assessment context

For the summative assessment, the rubric and standard stay identical for all learners, but the context, audience, and use case are personalised — one learner applies a framework to a public school, another to a corporate training environment. Same rigor, different context. This is an adaptive mechanism and an integrity mechanism at once, because a shared answer no longer fits either learner's context.

The eight mechanisms, together

Diagnostic-band routing · misconception-targeted remediation · prerequisite & bridge routing · milestone readiness gating · misconception blockers on submission · enrichment paths for secure mastery · Companion behaviour shifts by band (see the Companion) · personalised assessment context. Built incrementally — the obvious cases first, widened by real usage and metrics, with no gaps where the learner is left without guidance.