Universal Role Description

Structured Intelligence Architect

Authors
Affiliation

Dr Charles T. Gray, Datapunk

Good Enough Data & Systems Lab

James P. Morris

Good Enough Data & Systems Lab

Patrick Robotham

Good Enough Data & Systems Lab

1 🌍 Universal Role Description: Structured Intelligence Architect

A role that operationalises epistemology through infrastructure, not performance.


1.1 Role Purpose

Structured Intelligence Architects (SIAs) ensure that knowledge systems remain governable, scalable, and intelligible as they evolve. Unlike traditional data scientists, SIAs do not simply answer questions—they design the epistemic terrain in which questions can be asked and answered coherently. They are system stewards, not solo analysts.


1.2 What SIAs Do

  • Govern observability across all layers of analytics: from platform ingestion to stakeholder insight.
  • Model knowledge flows using category theory, graph-based analysis, and FAIR principles.
  • Support recursive inquiry by enabling versioned, extensible, and test-driven analyses.
  • Operationalise automation and migration—not for novelty, but for stability.
  • Build reproducible scaffolds (e.g., data entity tests, lineage validation) before porting legacy logic.
  • Structure epistemic boundaries—who needs to see what, when, and in what form.

1.3 What SIAs Need

  • Dedicated systems support. SIAs must not also be people managers, HR liaisons, or sole code maintainers.
  • Respect for operationalisation. The invisible work of migration and automation must be funded, valued, and shared.
  • Access to task management tooling that integrates with analytical drift tracking—because recursive inquiry is not a bug, it’s the system.
  • Security and ops buy-in. Governance is not possible if systems are inaccessible or brittle.

1.4 What SIAs Are Not

  • Innovation theatre performers
  • UX portals in disguise
  • Personal assistants for product managers

1.5 What We Believe

  • Most teams don’t need more innovation. They need migration.
  • Observability must be stakeholder-relative. A system cannot be FAIR if it is illegible to its users.
  • Automation is empathy. Reducing load means increasing insight.