Product · Pre-release
Lakewright reads the data nobody else can, and lands it governed.
Claims files, HL7 feeds, mainframe extracts, scanned documents, database change streams: identified, classified, and written to Apache Iceberg tables inside your own environment. An AI proposes the plan. A person approves it. Nothing leaves the building.
Formats other tools leave on the floor.
Self-hosted
Runs inside your environment. OCR runs on your own hardware, so documents with protected data never leave your network.
Governed
Every ingestion plan is proposed by an AI, approved by a person, and recorded. Nothing loads without that approval.
Ready for your warehouse's AI
Generates semantic layers for Snowflake Cortex, Databricks Genie, Microsoft Fabric and dbt, and ships an MCP server so assistants can ask what governed data exists.
The product has its own site.
Ingestion patterns, format-by-format briefs and the product's own contact form live at lakewright.io. Lakewright AI is the firm that builds it and puts it to work.