AivexaNewsSearch
AI news for builders and product teamsChecked every hour

How Block powers collaborative AI agents with Buzz and Unity Gateway

Collected Nov 12, 2026

Block has built an internal system called Buzz that puts employees and AI agents into a single shared workspace, with Databricks providing the underlying data and model access through a component Block calls Unity Gateway. The details come from a Databricks write-up describing how the setup is used inside Block.

The arrangement points to a pattern worth watching: rather than treating agents as standalone chat tools, Block is placing them inside the same environment where staff already collaborate, so humans and agents share context. Unity Gateway appears to act as the connective layer between that workspace and Databricks, which suggests the platform handles access to data and models rather than each agent wiring up its own connections.

The public material here is thin on specifics — there are no published numbers on adoption, latency, cost or the range of agent tasks in use — so the concrete claims should be read as a description of architecture and intent rather than measured results. This is my own reading of what has been shared.

Why it matters: teams evaluating agent deployments are increasingly asking how to give agents governed access to company data without rebuilding plumbing for each one; a shared gateway plus a common workspace is one answer, though this account offers no benchmarks to judge it by. For developers, the interesting design choice is centralizing access rather than embedding it per agent — inference, not a stated conclusion.

Read at Databricks

Based on reporting from the original publisher. Visit the source for full context and later updates.

Publisher excerpt

Inside Block, Buzz brings employees and AI agents into a shared workspace. With Databricks...