A fascination with how the world organizes itself
WorldOfTaxonomy started with a simple observation: every country, every industry, and every international body has its own way of classifying the same things - jobs, products, diseases, trades, education, risk. And anyone working across borders has to navigate all of them.
The problem that sparked this
If you have ever worked in global trade, you know the headache. Your product has an HS code for customs, an NAICS code for statistics, a CPC code for the UN, and a completely different classification in the country you are shipping to. A doctor writing a referral in Germany uses ICD-10-GM while their colleague in Australia uses ICD-10-AM. A recruiter posting the same role in the US, the EU, and India is dealing with SOC, ISCO, and NIC respectively.
These are not obscure bureaucratic artifacts. These systems shape how governments measure economies, how hospitals get paid, how trade flows across borders, and how researchers compare data across nations. They are the invisible infrastructure of the global economy.
Yet there was no single place to explore them side by side, see how they connect, or translate between them. That gap fascinated me.
An experiment in connecting standards
Each dot is a classification system. Lines are crosswalk connections between them. Click to explore.
WorldOfTaxonomy is an attempt to bring these systems together. Not to replace them - each one exists for a reason - but to connect them as peers in a single knowledge graph. One place where you can look up an NAICS code and see its equivalent in ISIC, NACE, ANZSIC, and a hundred other national systems. Where you can explore how medical codes map across ICD-10, ICD-11, LOINC, and SNOMED. Where occupation taxonomies from the US, EU, and the ILO are linked through crosswalk edges.
Today it connects over 1,000 classification systems, 1.2 million nodes, and 320,000+ crosswalk edges - spanning industry, medicine, trade, education, occupations, regulations, and more.
Why this matters in the age of AI
AI is fascinating. And one thing AI needs desperately is structured, reliable knowledge about how the real world is organized. When an AI agent processes a trade document, drafts a compliance report, or helps a researcher compare healthcare data across countries, it needs to navigate the same taxonomy maze that humans do.
A global standards connection system like WorldOfTaxonomy can be genuinely useful here - not just for humans browsing a website, but for AI systems that need structured crosswalk data through APIs and MCP servers. The goal is to make these taxonomies machine-accessible so that both humans and AI can work with them fluently.
This is an open experiment
WorldOfTaxonomy is an experimental effort, open source and evolving. The APIs and MCP servers are being published to empower anyone - researchers, developers, policy analysts, AI builders - who works across classification boundaries.
I would love for the community to use it, break it, and tell me what to improve. Whether you spot a wrong mapping, want a system added, or have ideas for how this could be more useful - your feedback is what makes this better.
Built with curiosity by Ram Katamaraja and the Colaberry AI team.