wtf is amundsen?

Amundsen is a search engine for data. Just like Google helps you find web pages, Amundsen helps data analysts, scientists, and engineers find the specific tables, dashboards, and datasets they need within their company's data warehouse or lake. It solves a real pain point: when you're working with data, you often don't know where the table you need lives, what columns it contains, or whether anyone else is using it. Amundsen finds that information for you in seconds. The way it works is straightforward. First, the system indexes all your company's data resources, tables from databases like Redshift, BigQuery, Snowflake, and many others, plus dashboards from tools like Tableau or Superset. It learns which tables are used most frequently and by whom. Then when you search, it ranks results the same way Google does: highly-used tables appear higher in the results than rarely-touched ones. You see not just the table name, but details like column names, descriptions, data types, sample statistics, and who last modified it. There's also a web interface where you can browse and explore. The project is built from separate but connected pieces. There's a frontend (a web interface), a search service (powered by Elasticsearch), a metadata service (using databases like Neo4j or Apache Atlas), and a data ingestion tool that pulls information from your existing systems. You can set it up to automatically refresh metadata from your warehouse on a schedule, often using Apache Airflow. This means Amundsen stays in sync with your actual data as it changes. Companies like Lyft, Square, and Edmunds use Amundsen to solve the "data discovery" problem at scale. A data analyst who wants to find customer transaction tables doesn't have to email the data team or dig through documentation anymore, they just search. It also helps with governance: by tracking who uses what data, teams can better understand dependencies and manage data quality. The tool supports most major database systems and works whether your data lives in the cloud or on-premises.