Skip to main content

Our Methodology

This fellowship is built on proven educational frameworks and assessment principles, ensuring you develop real capability, not just theoretical knowledge.

Bloom's Taxonomy - Progressive Skill Development​

Our curriculum follows Bloom's Taxonomy, progressing from foundational knowledge to advanced application:

  • Remember & Understand - Master core concepts and frameworks
  • Apply - Use knowledge in real-world scenarios
  • Analyze - Break down complex problems and evaluate solutions
  • Evaluate - Make judgments about when and how to use AI
  • Create - Build production-grade artifacts and systems

Each track level builds on the previous, ensuring you develop deeper capabilities as you progress. This structured progression means you're not just learning concepts in isolation - you're building a comprehensive skill set that compounds over time.

Knowledge Dimensions - Comprehensive Learning​

We address four critical knowledge dimensions:

  • Factual Knowledge - Understanding AI concepts, frameworks, and best practices
  • Conceptual Knowledge - Grasping relationships between ideas and systems
  • Procedural Knowledge - Knowing how to execute AI development and governance processes
  • Metacognitive Knowledge - Developing awareness of your own thinking and learning processes

This multidimensional approach ensures you can both execute and adapt your knowledge to new contexts. You'll understand not just what to do, but why it works and how to adapt it when circumstances change.

Artifact-Based Assessment - Real Capability Demonstration​

Rather than exams or theoretical exercises, you demonstrate capability through production-grade artifacts:

  • Real deliverables - Build actual frameworks, systems, and documentation
  • Reusable outputs - Create artifacts you can use in your work
  • Enterprise quality - Meet production standards, not just academic requirements
  • Portfolio building - Develop a body of work that demonstrates your capability

Assessment focuses on what you can create and deliver, not just what you can recall. Each artifact you produce becomes part of your professional portfolio, demonstrating real capability to employers and stakeholders.

Transfer & Judgment - Applied Intelligence​

Two critical capabilities distinguish expert practitioners:

  • Transfer - Apply learning to novel situations and contexts. You'll demonstrate ability to adapt frameworks and knowledge to new problems, showing you understand principles deeply enough to apply them flexibly.
  • Judgment - Make sound decisions under uncertainty. You'll show restraint, tradeoff awareness, and the wisdom to know when not to use AI - often the mark of true expertise.

These dimensions are explicitly assessed because they're what separate capable practitioners from those who can only follow recipes. In real-world AI work, you'll face situations that don't match textbook examples. Transfer and judgment are what enable you to navigate these challenges successfully.

Why This Methodology Matters​

This approach ensures you develop:

  • Practical capability - Not just knowledge, but the ability to apply it
  • Adaptability - Skills that transfer across contexts and evolve with technology
  • Professional judgment - The wisdom to make sound decisions in complex situations
  • Credible credentials - Badges backed by real artifacts and demonstrated capability

Our methodology is grounded in educational research and validated by decades of professional development practice. You're not just learning - you're building demonstrable, transferable capability that will serve you throughout your career.

Assessment Dimensions​

All assessments evaluate five universal dimensions across all tracks:

Judgment​

Quality of decisions under uncertainty - making sound choices when information is incomplete, showing restraint, and knowing when not to use AI.

Rigor​

Evidence-based reasoning - using data and evidence to support decisions, applying proper evaluation methodologies, and demonstrating systematic thinking.

Transfer​

Ability to apply learning to new contexts - adapting knowledge to new situations, generalizing from specific examples, and applying frameworks across domains.

Governance​

Awareness of risk, controls, and accountability - identifying and mitigating risks, implementing proper controls and guardrails, and ensuring accountability and auditability.

Tool & AI Mediation​

Understanding how AI tools augment human capability - knowing when to use AI tools vs. traditional methods, evaluating tool capabilities and limitations, integrating tools into workflows effectively, and understanding how tool-mediated cognition shapes thinking and decision-making. This dimension recognizes that modern AI work is inherently tool-mediated, requiring professionals to understand how tools shape capability, when tool mediation enhances vs. constrains work, and the implications for governance and accountability.

These dimensions ensure that capability is assessed holistically, not just on technical skills but on the professional judgment, governance awareness, and tool mediation understanding that make AI implementations successful.

Agentic stack reference (AI Capabilities Stack)​

The AI Capabilities Stack is the interactive source of truth: five tracks (T1-T5), eight agentic stack planes and five enterprise planes, weekly curricula, artifacts, Synthesis Principles cards, and two footer bars below the diagram.

Row 1 — Stack maturity (agentic / technical path): Integration / ESB → Data Engineering → RAG → Memory → AR/VR (emerging). Enabling: AI SDLC + Cost Mgmt (cross-cutting).

Row 2 — Enterprise planes (org and economics themes, not a second technical dependency chain): People & Leadership → Culture & Change → Governance & Risk → Strategy & Portfolio → Economics & ROI (same order and colors as enterprise rows in the diagram; square markers in the footer).

Synthesis Principles (diagram)​

  • Pedagogy layer (Fellowship): Bloom’s Taxonomy · Artifact-driven · Role-specific · Governance-first
  • Technical depth (Playbook): Production code patterns · Build-vs-buy · 2026 tooling · Plane architecture
  • Maturity dependency chain: Integration → Data Eng → RAG → Memory → AR/VR (emerging) · Enabling: AI SDLC + Cost Mgmt (footer row 1; row 2 lists enterprise planes)
  • Key differentiator: Memory ≠ RAG · Governance outside the agent · Planes as risk surface map

Eight planes (same wording as the diagram)​

  • UX - Where trust is won or lost. What users see, approve, and download. Maturity: borrowed surfaces → dedicated hub → hybrid.
  • Control - Policy, identity, tool registry, budget enforcement, progressive delivery. Lives outside the agent loop - deterministic, not prompt-based.
  • Runtime - Multi-lane execution: frontdoor / durable worker / sandbox / GPU inference. Each lane has its own latency, isolation, and cost contract.
  • Context & Memory - Memory enables cognition - RAG only optimizes retrieval. Governed memory frameworks (example: open-source Astrocytes; see Glossary: Astrocytes (framework)) implement retain/recall/reflect with policy—not “embedding search only.” Four networks: World, Experience, Opinion, Observation. TEMPR (Temporal Entity Memory Priming Retrieval) — structured retrieval used by Hindsight. Vectors for similarity retrieval; graph stores for relational structure; hybrid patterns where both apply.
  • Data - Multi-tenant data ownership, behavior schemas, tool I/O contracts. Schema as source-of-truth for agent behavior.
  • Integrations - Web search, scraping, external APIs, drift management. Protocol convergence: MCP → ACP → A2A.
  • Security - Identity, authorization, I/O guardrails. Compliance: CA SB 243/AB 489. Zero-trust workload identity.
  • Observability - Traces, metrics, eval, cost attribution. OTel GenAI semantic conventions for vendor portability. Three eval strategies.

Ready to experience this methodology firsthand? Get in touch to learn more or apply.

Previous: Applied AI Fellowship | Next: T1: Executive