Agentic Infrastructure for the Autonomous Enterprise on Google Cloud (AIAEGC) – Outline

Detailed Course Outline

Module 1 - The Shift to Agency

Topics:

  • 1. The Pillars & Governance
  • 2. The Architectural Friction Forces
  • 3. The Autonomous Maturity Scale

Objectives:

  • Analyze the "Agency Gap" by diagnosing the functional intersection of Reasoning, Memory, and Tools with Data Governance to move from Track A (Chat) to Track B (Workers).
  • Diagnose the four technical frictions (Integration, Statelessness, Latency, and Governance) that prevent AI pilots from scaling into production.
  • Evaluate infrastructure readiness using the L1–L5 Maturity Scale to prioritize "Paved Road" investments for Level 4+ autonomy.

Activities:

  • 1 Use Case
  • 2 Case Studies
  • 1 Demo

Module 2 - Building the "Paved Road"

Topics:

  • 1. Reference Stack & Tool Archetypes
  • 2. The Memory Decision Guide
  • 3. Multi-Agent Orchestration Patterns
  • 4. The Paved Road Lifecycle

Objectives:

  • Apply the Vertex AI SDK and Vertex AI Reasoning Engine to standardize agent deployment and manage persistent conversation state.
  • Evaluate the trade-offs between AlloyDB and Vertex AI Vector Search to select the optimal storage layer for metadata-heavy vs. high-scale agents.
  • Apply specific orchestration patterns (Hub-and-Spoke, Linear Relay, or Parallel Critic) to manage complex, multi-departmental goals.
  • Design an agentic deployment arc from Sandbox to Certified production to ensure infrastructure precedes autonomous action.

Activities:

  • 4 Demos

Module 3 - The Autonomous Perimeter

Topics:

  • 1. Threat Modeling for Agentic Systems
  • 2. Identity Hierarchy & Credentials
  • 3. Defending the Boundary: Model Armor
  • 4. Responsible AI & Human-in-the-Loop

Objectives:

  • Apply agentic threat modeling to identify and mitigate risks like Indirect Prompt Injection and Tool-Chaining exploits.
  • Apply a three-layer identity model using Workload Identity Federation to ensure "Least Privilege" for autonomous workers.
  • Apply Model Armor as a real-time security proxy to filter malicious inputs and redact sensitive output data.
  • Analyze Responsible AI production requirements to embed accountability, traceability, and "Human-in-the-Loop" checkpoints within the Autonomous Perimeter.

Activities:

  • 1 Use Case
  • 4 Demos

Module 4 - Sustaining Autonomy

Topics:

  • 1. Infrastructure ROI
  • 2. The GenAIOps Lifecycle
  • 3. The Innovation Harvest

Objectives:

  • Analyze platform ROI by shifting from vanity metrics to Infrastructure Leverage Ratios and Component Reusability to prove the value of the "Paved Road."
  • Apply a continuous feedback loop using Golden Datasets and reasoning traces to detect and remediate "Reasoning Drift."
  • Apply the "Innovation Harvest" methodology to scale successful siloed tools into global, certified Gemini Enterprise assets.

Activities:

  • 1 Use Case
  • 1 Demo

Module 5 - Summary and Quiz

Topics:

  • Review of Core Concepts

Objectives:

  • Evaluate understanding of core course concepts through scenario-based questions.

Activities:

  • 5 scenario-based multiple choice questions.