irreducibly-human

Graduate course syllabus for engineers and technical practitioners at Northeastern University who use AI tools in design work. Covers the complete AImagineering pipeline — Empathize, Define, Ideate, Prototype, Test, Commit — with explicit focus on the human judgment calls AI cannot perform, culminating in a Commit document the student can stand behind.

Graduate course syllabus for professionals at Northeastern University who use AI tools in their work and need a framework for evaluating what they are working with. Builds the complete Five Modes architecture and nine-pillar AI fluency framework — the series entry point for Irreducibly Human, with no technical prerequisites.

Graduate course syllabus for engineers and applied technical practitioners at Northeastern University who use data to make decisions and need the identification layer that causal AI tools cannot perform. Covers DAG construction, confounders, mediators, colliders, the backdoor criterion, and sensitivity analysis — culminating in a complete, defensible causal analysis plan for the student

Graduate course syllabus for engineers and applied practitioners at Northeastern University who need the validation layer that AI systems cannot perform on themselves. Builds computational frameworks for systematic AI doubt: bias detection, explainability critique, adversarial testing, causal validation, and the metacognitive infrastructure to know when an AI output is wrong before recomputing it.

Graduate course syllabus for engineers and AI-adjacent professionals at Northeastern University who can operate AI tools fluently but cannot yet explain rigorously why their outputs should be trusted. Develops the five supervisory capacities no algorithm possesses — plausibility auditing, problem formulation, tool orchestration, interpretive judgment, and executive integration — culminating in an adversarial Plausibility Audit of the student

Graduate course syllabus for engineers and technical practitioners at Northeastern University who work with systems that encode values and have never been asked to make someone else feel morally implicated by something they built. Students design, build, and audit a web-based game whose ethical framework is identifiable from the mechanics alone — culminating in a gap analysis naming what an AI Ethical Auditor cannot find because finding it requires a player.

Master reference document for the Irreducibly Human curriculum series. Covers the seven-tier intelligence taxonomy, complete course inventory, production pipeline, and open action items as of March 2026.