Back office ops · Production

Assembled builds a company-wide AI operating system with Dust, achieving 95% internal adoption across 120+ employees

The problem

Assembled's hypergrowth exposed severe knowledge fragmentation: company information was scattered across Google Drive, Slack, Notion, Linear, Cursor, Snowflake, and more, search was broken, rapid product releases made knowledge quickly obsolete, and individual AI tool usage was siloed with no shared workflows.

First attempt

Individual automation setups using Relay, Zapier, or personal Claude MCP configurations did not scale because each workflow was tied to a single employee's account and required technical setup most staff could not do.

Workflow diagram · grounded in source
1
CEO pilots Dust in Slack
Trigger
Assembled's CEO began experimenting with Dust inside a Slack channel to test small internal workflows.
▾ source quote
“Ryan began experimenting with Dust inside a Slack channel, using it to test small internal workflows”
2
Unified cross-tool search
Ai action
Dust gave employees one place to search across all their tools with a single query.
▾ source quote
“With a single query, teams could pull information from: Notion, Google Drive, Slack, Gong, Salesforce”
3
Teams build specialized agents
Ai action
Teams built specialized Dust agents for their own workflows across all company departments.
▾ source quote
“Marketing built competitive intelligence and content automation agents. Sales created follow-up and account research agents. Customer Success built Zendesk and product feedback agents. Engineering added code-base search and SQL query agents. Product automated changelog and release mapping tasks”
4
AI Collective knowledge sharing
Feedback loop
Assembled created an internal AI Collective, a monthly gathering where employees share agents they've built and collaboratively troubleshoot new ideas.
▾ source quote
“Assembled created an internal AI Collective, which is a monthly gathering where employees share agents they've built and collaboratively troubleshoot new ideas”
5
Employee self-serve retrieval
Output
Employees now self-serve information through Dust rather than interrupting colleagues to locate documents.
▾ source quote
“It used to be: ask marketing or CS where something is. Now it's: ask Dust. Everyone can self-serve”
Reported outcome

Dust achieved 95% internal adoption across 120+ employees without extensive training, saved hundreds of hours through unified search and AI agents, reduced cross-team interruptions so employees self-serve information, and accelerated new-hire onboarding.

Reported metrics
Internal adoption rate95%
Hours saved overallhundreds of hours
Hours saved from search alonehundreds of hours
Reported stack
DustZendeskLinearSnowflakeRelayZapier
◆ Does this fit your context?

Compare to your context

Tell us your scale, team, and constraints. We'll show what changes at your size, what fails at your scale, and whether this case is a fit, needs adaptation, or won't scale to you. Free demo, no signup.

Compare to your context →
~30 seconds · free
Source
https://dust.tt/customers/part-1-assembled-ai-operating-system
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Dust achieved 95% internal adoption across 120+ employees without extensive training, saved hundreds of hours through unified search and AI agents, reduced cross-team interruptions so employees self-serve information,…

What tools did this team use?

Dust, Zendesk, Linear, Snowflake, Relay, Zapier.

What results were reported?

Internal adoption rate: 95%; Hours saved overall: hundreds of hours; Hours saved from search alone: hundreds of hours (source-reported, not independently verified).

What failed first in this deployment?

Individual automation setups using Relay, Zapier, or personal Claude MCP configurations did not scale because each workflow was tied to a single employee's account and required technical setup most staff could not do.

How is this back office ops AI workflow structured?

CEO pilots Dust in Slack → Unified cross-tool search → Teams build specialized agents → AI Collective knowledge sharing → Employee self-serve retrieval.

Related back office ops cases

WHAT TO DO WITH THIS

Now compare it to your context

This case is one data point. Whether its pattern fits you depends on your volumes, your stack, and your exception load — that comparison is the step no case study can do for you.

Browse the Back office ops cases →