The diagnostic engine that shows where your judgment breaks.
Bayesian Knowledge Tracing. Graph-native. Issues verifiable credentials.
P(MASTERY) OVER TIME · BKT MODEL
spent on training annually
of senior leaders say companies lack skills
Most of that $400B targets guesses, not diagnosed gaps. The Scaffold changes that.
A Bayesian model that updates its estimate of your competency with every assessment response. Not a quiz. Not a self-assessment. A calibrated diagnostic.
Scaffold reads your role + domain
The diagnostic engine loads your certification domain and role context from the knowledge graph. No guesswork — structured assessment from first response.
BKT runs, HAS_GAP is written
Bayesian Knowledge Tracing updates your competency estimate with every response. When mastery drops below 0.60, HAS_GAP is written to Neo4j.
Mirror Moment with credential
Your first gap report, in plain language. The Scaffold issues an issue_diagnostic_vc as a W3C VC 2.0 verifiable credential.
Unified vector + graph — semantic search and graph traversal in a single substrate.
Below threshold triggers HAS_GAP. APPLY-face assessment triggers HAS_MASTERED.
W3C VC 2.0 + SD-JWT. Selective disclosure — show pass/fail without revealing scores.
No competitor has a Bayesian-calibrated gap diagnostic connected to a content production engine.
What is Bayesian Knowledge Tracing?
How accurate are the diagnostics?
What is a verifiable credential?
Can the Scaffold work standalone?
Standalone or composable
Use the Scaffold alone for gap diagnostics, or connect it to the Cubelets. When a gap is diagnosed, Course Factory routes targeted content automatically. Both compound.
Start free →