Service principal + Unity Catalog
Reads Delta tables and views through a service principal with Unity Catalog permissions limited to approved catalogs and schemas.
Integrations — Databricks
Vericent works with Databricks Lakehouse data in Delta tables, governed by Unity Catalog, from batch history to streaming transactions.
01 — Rationale
Databricks customers often hold the richest, rawest data: logs, events, sensor feeds and transactions alongside curated tables. That is exactly what anomaly detection needs. Vericent applies its behavioural models to this data without forcing you to move it to another platform, and complements the models your own data science team builds.
02 — How it connects
Vericent connects through the path that fits your architecture, security policies and budget.
Reads Delta tables and views through a service principal with Unity Catalog permissions limited to approved catalogs and schemas.
Option to receive data through Delta Sharing, without building new export pipelines.
Supports both batch (historical baselining and back-testing) and streaming sources for near real-time scoring.
Writes scores, alerts and explanations back to Delta tables for use in notebooks, dashboards and downstream jobs.
Use Vericent scores as features in your own models, or vice versa.
03 — What you can run
All Vericent solutions. Databricks customers often start with:
04 — Outcomes
Outcome 1
Use raw and curated lakehouse data together.
Outcome 2
Governance stays in Unity Catalog.
Outcome 3
Explainable, production-ready detection without building it from scratch.
Outcome 4
Your data science team keeps full visibility of inputs and outputs.
05 — Stakeholders
Chief Data Officers, data platform owners, and data engineering and data science teams running Databricks.
Start with a conversation about your data estate and the integration path that fits.