architecture-handbook

AI Learning Path for Technical Architects

A structured learning guide for technical architects looking to understand modern AI/ML capabilities, agentic workflows, and architecture automation patterns.

Disclaimer: This learning path been created from a learning path based on human intiution. Learning notes were fed into Claude Code, it was tasked with producing this learning path based on those notes. The path has been reviewed by a human, any fully AI generated content that has not been validated has been marked with (AI Generated). Additionally this guide contains hypotheses about how AI Agents could be used to aid Architects, these have been marked with (Hypothesis).

Who This Is For

This learning path is designed for technical leaders who need to:

Understand how AI can augment development workflows
Evaluate AI tooling and patterns for their teams
Design systems that integrate with AI/ML capabilities
Make informed build vs. buy decisions for AI-powered solutions

How to Use This Guide

Each topic includes:

Video resources (primary learning material)
Key concepts (80/20 insights)
How it works (technical explanations with diagrams)
Common approaches (trade-offs, no prescriptive “winners”)
Try it yourself (hands-on experiments where applicable)

Capability Matrix

Topic	Difficulty	Key Architectural Takeaway	Section
Agentic AI Evolution	Introductory	How LMs evolved from simple prompts → RAG → autonomous agents	01-foundations
AI Safety & Control	Intermediate	Understanding adversarial AI, monitoring, and the capacity gap	01-foundations
MLOps Principles	Introductory	ML systems are 10% ML, 90% good engineering	01-foundations
Prompt Engineering	Introductory	CoT, few-shot examples, systematic evaluation	02-core-patterns
RAG Architecture	Intermediate	Vector stores, chunking, retrieval for domain-specific knowledge	02-core-patterns
Agentic Workflows	Intermediate	ReAct, reflexion, tool use, multi-agent collaboration	02-core-patterns
Vibe Engineering	Intermediate	The paradigm shift in AI-assisted development	03-development-workflows
Spec-Driven Development	Intermediate	Context compression: maintaining understanding at AI generation speed	03-development-workflows
Agent Skills Framework	Intermediate	Anthropic Skills: procedural vs. declarative AI capabilities	03-development-workflows
Context Management	Introductory	Giving LMs the right context to reduce hallucinations	03-development-workflows
ADR Automation	Advanced	Using agent skills for architecture governance	04-governance-automation
Security Automation	Advanced	Threat modeling and IaC security auditing with agents	04-governance-automation
Integration Contracts	Advanced	API contract validation and schema governance	04-governance-automation
Architectural Drift Prevention	Advanced	Preventing the comprehension gap and measuring drift with ADI	04-governance-automation

Suggested Learning Paths

Path 1: Foundations First (Recommended for teams new to AI)

01-foundations/
  → agentic-ai-evolution.md
  → mlops-principles.md

02-core-patterns/
  → prompt-engineering.md
  → rag-architecture.md

03-development-workflows/
  → vibe-engineering.md
  → context-management.md

Path 2: Hands-On Development (For teams already using AI tools)

03-development-workflows/
  → vibe-engineering.md
  → agent-skills-framework.md
  → context-management.md

02-core-patterns/
  → prompt-engineering.md
  → agentic-workflows.md

Path 3: Architecture Governance (For teams experienced in leveraging AI tools)

01-foundations/
  → ai-safety-control.md

02-core-patterns/
  → agentic-workflows.md

04-governance-automation/
  → adr-automation.md
  → security-automation.md
  → integration-contracts.md

By Role

Enterprise Architects: Focus on 04-governance-automation
Solution Architects: Start with 02-core-patterns + 03-development-workflows
Security Architects: Prioritise ai-safety-control.md + security-automation.md
Team Leads: Begin with vibe-engineering.md + context-management.md

By Use Case

Evaluating AI tools: Start with agent-skills-framework.md + vibe-engineering.md
Building AI-powered systems: Focus on 02-core-patterns section
Governance & safety: Navigate to 01-foundations/ai-safety-control.md + 04-governance-automation
Team enablement: Begin with 03-development-workflows section

Resources

Video Library - All curated video content organised by topic
Tool Comparison - Claude Code, Cursor, MCP servers, and more
Glossary - Key terms and concepts

Contributing

This is a living learning guide. If you’ve discovered valuable resources or have insights to add, contributions are welcome.