What This Is

Semantic Stack is a resource for designing better data systems in the age of AI.

It combines:

  • Best practices → practical, downloadable guides (Markdown) for data modeling and semantic layers
  • Strategy → clear thinking on how data architectures evolve with AI
  • Tools → a lightweight explorer to understand data before modeling

The Problem

AI is making data work faster—but not better.

Most teams still struggle with:

  • unclear data structures
  • unreliable joins and metrics
  • fragmented business logic
  • low trust in outputs

AI doesn’t fix this. It amplifies it.

The Approach

Semantic Stack is built on a simple principle:

You can’t model, govern, or use data with AI until you understand it.

Three focus areas:

  • Data Understanding → profiling, relationships, structure
  • Semantic Modeling → consistent metrics and business logic
  • AI Context → adding meaning, documentation, and relationships for reasoning

What You’ll Find Here

  • Downloadable best practices (MD) Clear, reusable frameworks for:
    • data modeling
    • semantic layers
    • data quality and joins
  • Strategy & perspective How to design data stacks that work with AI:
    • semantic layer as control plane
    • measurement vs meaning (metrics vs relationships)
    • reducing complexity and increasing trust
  • Tools A local, in-browser explorer to:
    • profile datasets
    • detect joins and keys
    • generate context for modeling

Point of View

  • Semantic layers are more important with AI, not less
  • Data modeling is shifting from SQL writing → data understanding
  • The biggest bottleneck is trust, not computation

Who This Is For

Data professionals who want to build:

  • reliable metrics
  • clear data models
  • AI-ready data systems