Learning Hub

Learning Hub is structured, foundational content for architects and technology leaders: AI concepts explained for architects, architecture fundamentals, learning paths, framework comparisons, and a working glossary. The goal is not to turn every architect into an ML researcher — it's to learn enough to understand, design, evaluate, and lead.

Path 1: The AI Architect's Blueprint

A five-part foundation series, in order — start here if you're building AI architecture knowledge from the ground up.

  1. LLM & Generative AI Foundations
  2. RAG Systems & Vector Databases
  3. Agentic AI: Agents, MCP & Multi-Agent Systems
  4. Enterprise AI Architecture & LLMOps
  5. AI Governance, Risk & Domain Applications

Coming Soon

More learning paths, standalone explainers, and comparisons are in development and will publish on a rolling basis:

Additional Learning Paths

Follow-on paths for after you've completed Path 1.

Planned

Path 2: AI Platform Engineering Fundamentals

The operational layer, explained conceptually — gateways, LLMOps, and observability, and why each one exists, before you ever open the Playbooks.

Planned

Path 3: AI Governance & Risk Fundamentals

The regulatory and compliance landscape for AI, explained for architects — risk classification, accountability, and the frameworks shaping enterprise AI policy.

Concept Explainers

Short, standalone "what is X" pieces — the fastest way to get oriented on one idea at a time.

Planned

What Is an AI Agent?

Agents vs. chatbots vs. workflows — what actually makes something "agentic," and where the term gets misused.

Planned

Tokens, Context Windows & Embeddings, Explained

The three concepts every architect needs before any of the rest of this site makes sense.

Planned

What Is Model Context Protocol (MCP)?

A plain-language explanation of MCP, what problem it solves, and how it's different from a plain API integration.

Planned

Fine-Tuning vs. RAG vs. Prompt Engineering

Three different ways to customize a model's behavior, and how to tell which one you actually need.

Planned

What Is an AI Gateway?

Why enterprises are putting a gateway in front of every model call, and what it actually does.

Planned

SLMs vs. LLMs: When On-Prem/Edge Actually Makes Sense

Why smaller, cheaper models running on-prem or at the edge are winning specific use cases away from large cloud LLMs.

Framework & Tool Comparisons

Side-by-side educational comparisons — lighter than a formal vendor-selection process.

Planned

Agent Framework Comparison: LangChain vs. LlamaIndex vs. Semantic Kernel vs. AutoGen

A side-by-side look at how the major agent frameworks differ, without the vendor-selection scorecard treatment.

Planned

Vector Database Comparison: Pinecone vs. Weaviate vs. pgvector vs. Milvus

What actually differs between the major vector database options, in plain terms.

Glossary

A single, living reference — not a series, just kept up to date.

Living Doc

The AI Architecture Glossary

A living reference of AI and architecture terms used across this site, added to over time.

Coming Soon

More learning paths, standalone explainers, and comparisons are in development and will publish on a rolling basis:

Additional Learning Paths

Follow-on paths for after you've completed Path 1.

Planned

Path 2: AI Platform Engineering Fundamentals

The operational layer, explained conceptually — gateways, LLMOps, and observability, and why each one exists, before you ever open the Playbooks.

Planned

Path 3: AI Governance & Risk Fundamentals

The regulatory and compliance landscape for AI, explained for architects — risk classification, accountability, and the frameworks shaping enterprise AI policy.

Concept Explainers

Short, standalone "what is X" pieces — the fastest way to get oriented on one idea at a time.

Planned

What Is an AI Agent?

Agents vs. chatbots vs. workflows — what actually makes something "agentic," and where the term gets misused.

Planned

Tokens, Context Windows & Embeddings, Explained

The three concepts every architect needs before any of the rest of this site makes sense.

Planned

What Is Model Context Protocol (MCP)?

A plain-language explanation of MCP, what problem it solves, and how it's different from a plain API integration.

Planned

Fine-Tuning vs. RAG vs. Prompt Engineering

Three different ways to customize a model's behavior, and how to tell which one you actually need.

Planned

What Is an AI Gateway?

Why enterprises are putting a gateway in front of every model call, and what it actually does.

Planned

AI Pair Programming vs. Agentic Coding: A Spectrum

Where autocomplete, Copilot-style suggestions, and autonomous coding agents each sit — and what changes as autonomy increases.

Planned

Context Engineering for Code

What an AI coding tool actually needs to see — repo context, specs, style guides — to produce usable output, and why prompting for code differs from prompting for prose.

Framework & Tool Comparisons

Side-by-side educational comparisons — lighter than a formal vendor-selection process.

Planned

Agent Framework Comparison: LangChain vs. LlamaIndex vs. Semantic Kernel vs. AutoGen

A side-by-side look at how the major agent frameworks differ, without the vendor-selection scorecard treatment.

Planned

Vector Database Comparison: Pinecone vs. Weaviate vs. pgvector vs. Milvus

What actually differs between the major vector database options, in plain terms.

Planned

AI Coding Assistant Landscape

Copilot, Cursor, Claude Code, Windsurf, and similar tools compared on autonomy level, IDE integration, and enterprise controls.

Planned

AI Code Review Tool Comparison

Where automated review tools fit alongside human review, and what they catch versus miss.

Glossary

A single, living reference — not a series, just kept up to date.

Living Doc

The AI Architecture Glossary

A living reference of AI and architecture terms used across this site, added to over time.

AI-Native Development Practices

Practical guidance for working with AI day to day in the engineering workflow.

Planned

Best Practices for Reviewing AI-Generated Code

What to check for that differs from reviewing human-written code.

Planned

Best Practices for Working with Coding Agents

Scoping tasks, giving feedback, and knowing when to take back manual control.