Answers are grounded in this paper and cite their sources.
Demonstrates the importance of operational infrastructure for dropout prediction in higher education, suggesting improved institutional supports are needed.
Educational dropout in higher education is widely framed as a prediction problem. This paper argues theframing is premature. Without ordered operational infrastructure that records, tracks, and audits studentadvisor interactions, the data required for credible prediction does not exist, and the regulatory obligationsthat justify intervention cannot be operationally executed. PAEChakas is presented as the infrastructure layerthat must precede ML — a multi-tenant SaaS for institutional student support built on a hexagonal architecture, a pure Rust backend (Axum, SQLx, PostgreSQL), and a Silo + Control Plane deployment model. Wedescribe the system intent and the public Architecture Decision Records (ADRs) that constitute its designsurface, and we justify deferring the predictive ML module to a later phase governed by an inductive —not statistical — model. The decision reflects two empirical facts: dropout events are statistically atypicalwithin an active cohort, undermining majority-trained classifiers; and the Peruvian regulator SUNEDU already mandates institutional student support, making operational executability the binding constraint, notpredictive insight. The paper closes the trilogy started by AnimaLab (Paper 1) and AnimaCore (Paper 2) bydemonstrating how spec-first architecture meets a real regulated domain
No takes yet. Share an insight, caveat, or question.
Hendrik Hendrik Ordoñez Bullon (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: