#121
B+
Rank #98
Lightdash
An open-source BI tool that layers a semantic modeling layer and exploration UI directly on top of dbt projects. Standout trait is treating dbt models as the single source of truth for metrics, avoiding metric-definition drift.
70.8
/ 100
⚙ Full Stack
TypeScript monorepo (Node backend + React frontend) backed by Postgres for app metadata, querying the user's own warehouse (Postgres, Snowflake, BigQuery, etc.) for actual data via dbt-defined models. Official docker-compose stack bundles the app and its Postgres.
📈 Scaling Analysis
Scales well for query-serving since the heavy analytical lifting is pushed down to the underlying warehouse rather than done in-app; the Lightdash tier itself is a fairly thin, horizontally-replicable layer. Metadata Postgres is rarely the bottleneck — warehouse query performance is.
🚀 Running on Nexlayer
Two-pod minimum — app and its metadata Postgres (`postgres.pod:5432`) — plus network access out to whatever warehouse holds the actual analytical data (not itself hosted on Nexlayer typically). `SITE_URL` must be set to the Nexlayer `<% URL %>` for auth callbacks and shared dashboard links to work.