Integrations/Databricks

Control what your AI agents do in Databricks.

Let your team build AI agents for Databricks. Bollard AI controls what each one can do, area by area.

Request early accessSee how it works →
Databricks
Data Agent
SQL warehouse Data areaOffReadFull use
Catalogs Data areaOffReadFull use
Clusters Data areaOffReadFull use
Start clusters and grant access
What agents can do

Set what each agent can do across your workspace.

Each area of Databricks, like warehouses, catalogs, or clusters, gets one of three levels: Off, Read, or Full use. Anything not allowed is denied.

Off
No access
The agent can't see that part of your workspace.
Read
Read-only
The agent can read catalogs and tables in that area. It can't change anything.
Full use
Query and run
The agent can run SQL and jobs in that area.

Some actions stay off until explicitly turned on

Each one runs up compute cost, removes work for good, or hands out access, so it could override the limits set elsewhere.

Starting and stopping clustersDeleting jobs or cancelling runsGranting or revoking access
Set up & manage

Connect once. Manage everyone from there.

Connect once, then control access person by person. No new keys to hand out.

Step 1 · Set up

Connect once, company-wide

An admin connects your company's Databricks a single time. Bollard AI lays out the areas of your workspace, and anything you don't allow stays out of reach for every agent.

Step 2 · Manage

Control access per person and agent

From one dashboard, set what every person and their agents can reach. Change it any time, with no new keys for anyone.

OffNo access
ReadRead-only
Full useRead, query, and run jobs in a single area
Example

A data analytics agent.

A data agent runs SQL on a warehouse and can read your catalogs, but it can't start clusters or grant access.

SQL warehouse Data areaOffReadFull use
Catalogs Data areaOffReadFull use
Clusters Data areaOffReadFull use

Manage access

ALLOWEDData Agent ran SQL on the SQL warehouse
ALLOWEDData Agent read a record in Catalogs
BLOCKEDData Agent tried to open Clusters, denied by default

Monitor activity

The security model

Top-tier security.

Bollard AI is access control for AI agents. It sits between your agents and Databricks, so they never hold your Databricks keys and never reach Databricks directly.

Everything runs through Bollard AI

Agents send their requests to Bollard AI, not to Databricks. Bollard AI checks each one against your rules and passes only the allowed requests through.

Denied by default

Every area starts off-limits, and starting clusters or granting access is held back on top of that. An agent gets in only where you've allowed it, and anything Bollard AI doesn't recognise is refused and held for review, never waved through on a guess.

Your keys stay sealed

Your Databricks keys are sealed in a vault unique to your company and never shown again. Bollard AI unlocks them only for the instant an allowed request needs it.

Activity you can read

Every attempt is recorded in plain English: who or which agent, what they tried, which warehouse or table, and the decision. The record is readable and can't be edited after the fact.

Questions

Databricks, answered.

Does connecting Databricks hand Bollard AI your whole workspace?+
Only the areas you allow. An admin connects your Databricks, and you decide which areas, like a warehouse or catalogs, each agent can reach. Bollard AI doesn't browse or copy your workspace. It sits in front of it and checks every request the agents make, passing or blocking each one.
Can an agent run up compute cost or grant access?+
Not unless you turn it on. Starting and stopping clusters, and granting or revoking access, are held off by default, so an agent can query where you allow without ever running up compute or handing out access, until you explicitly allow it.
Can an agent see more than the person who owns it?+
No. An agent reaches only what its owner reaches. An explicit Off always overrides an inherited permission.
Where do our Databricks keys live?+
Sealed in an encrypted vault, behind a key unique to your company. It's never shown again once connected, and Bollard AI unseals it only for the instant an allowed request needs it.
What happens when an agent tries something it's not allowed to do?+
Bollard AI blocks the request before it reaches Databricks, and the blocked attempt lands in the activity record with the reason.
How do we connect Databricks?+
An admin connects the company's Databricks once. Bollard AI seals the connection and lays out the areas of your workspace to set rules for. From then on, all management happens in the Bollard AI dashboard.
Early access

Put your team's agents to work in Databricks, without the risk.

We're onboarding our first development partners now. Tell us about your setup and we'll be in touch.

Request early accessAll integrations →