Lightdash
BI platform that turns your dbt or YAML metric definitions into dashboards, AI agents and embedded analytics.
Dashboards, AI agents and embedding from one governed metrics layer, managed through Git and CI.
Lightdash is a business intelligence platform where metrics, joins and permissions are defined once in a context layer, then reused by dashboards, AI agents and embedded charts. It suits data teams that already work in dbt or Git and want analytics changes reviewed like code.
What it is
You define metrics, dimensions, joins and access rules in dbt projects or standalone Lightdash YAML pointed at your warehouse. Charts and dashboards are stored as files, so you can preview them from the CLI, validate them in CI and review them in pull requests. Business users then explore dashboards or ask questions in plain English through AI agents.
Who it's for
It fits analytics engineers and data teams that own a warehouse such as BigQuery, Snowflake, Redshift, Databricks or Postgres and want governed self-serve BI. It is a poor match if you have no warehouse or no one comfortable with dbt, YAML and Docker or Kubernetes.
Key features
Context layer for metrics
Define metrics, joins, descriptions, caching and access rules once and reuse them everywhere.
BI as code
Charts and dashboards live as files you can preview, validate in CI and review in pull requests.
AI agents on governed data
Agents answer from your context layer, respect permissions and return inspectable queries.
Data Apps from a prompt
Build custom reports, workbooks and forecasting tools with your permissions and auth already applied.
Embedded analytics SDK
Embed dashboards, agents and apps in your product with row-level security and user attributes.
CLI and MCP server
Coding agents can install Lightdash skills, preview changes and validate the project from a terminal.
Up and running in 3 steps
Our honest verdict
What's great
- Metrics are defined once and reused by dashboards, AI agents and embedded charts, so numbers stay consistent.
- Works from dbt projects or standalone YAML, so dbt-centric teams do not rebuild their models.
- Preview and validate commands plus Git integrations let you review analytics changes like application code.
- Warehouse adapters cover BigQuery, Snowflake, Redshift, Databricks, Postgres, Trino, ClickHouse and more.
Where it falls short
- Self-hosting needs Docker or Kubernetes and a PostgreSQL backend, and the README points you to a production deployment checklist.
- Enterprise features such as the ee directory code need a commercial licence key on self-hosted setups.
- The README recommends Lightdash Cloud for AI agents and Data Apps, so the self-hosted core is the narrower product.
- It assumes a modelled warehouse and some dbt or YAML skill; it is not a drag-and-drop tool for raw spreadsheets.
Pricing & license
The core platform is MIT-licensed and free to self-host. Lightdash's pricing page (https://www.lightdash.com/pricing) shows Cloud Pro from $3000 per month with no per-seat pricing, usage-based AI features, and a custom-priced Enterprise plan. Embedding is an add-on priced per chart load.
Open remote AI-training roles
870 open remote AI-training roles, updated November 14, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
Does Lightdash need dbt?
No. The README says you can use dbt projects or standalone Lightdash YAML pointed at your warehouse, though the dbt integration is a headline feature.
Can I self-host Lightdash?
Yes. The README describes running it with Docker or Kubernetes, with Helm charts available. Enterprise features on a self-hosted install need a licence key from the Lightdash team.
Which databases does it connect to?
The README lists adapters for BigQuery, Snowflake, Redshift, Databricks, Postgres, Trino, ClickHouse and more.
