Sales ops · Production

Databricks builds internal Field AI Assistant on Mosaic AI to transform seller experience with GenAI agents

The problem

Databricks sellers are overwhelmed by the volume of information spread across siloed applications and need unified access to account, opportunity, and use-case data in their CRM, customer market insights in the lakehouse, and sales content such as enablement playbooks, competitive collateral, and product knowledge base articles.

Workflow diagram · grounded in source
1
Seller query or seeded prompt
Trigger
The field assistant responds to seeded prompts based on user and page context and provides a chat-like interface for open-ended queries.
▾ source quote
“The field assistant responds to seeded prompts based on user and page context and also provides a chat-like interface for open-ended queries on the above-mentioned datasets”
2
LLM interprets request intent
Ai action
LLMs use context to interpret the intent of an ambiguous human request and convert it into something more deterministic.
▾ source quote
“LLMs have now given us the ability to use context to interpret the intent of a request and convert this into something more deterministic”
3
Driver agent retrieves and processes data
Ai action
The field AI assistant has one driver agent and multiple tools and functions that perform the deterministic processing.
▾ source quote
“The field AI assistant has 1 driver agent and multiple tools and functions that perform the deterministic processing”
4
Automated action orchestration
Output
The assistant orchestrates actions by automating repetitive manual administrative tasks.
▾ source quote
“orchestrate actions by automating repetitive manual administrative tasks”
5
AI Gateway guardrails and monitoring
Validation
The Agent Bricks AI Gateway provides access controls, rate limiting, payload logging, and guardrails filtering for system inputs and outputs.
▾ source quote
“The Agent Bricks AI Gateway provides access controls, rate limiting, payload logging, and guardrails (filtering for system inputs and outputs). The gateway gives the user constant monitoring of running systems to monitor for safety, bias, and quality”
Reported outcome

The Field AI Assistant helps sellers retrieve information, distill insights, and automate repetitive manual tasks, with the stated goal of significantly transforming every aspect of the seller journey and improving sales velocity.

Reported metrics
Seller journey transformationsignificantly transform and assist every aspect of the seller journey
Sales velocityimproving sales velocity
Reported stack
Databricks Mosaic AIAgent Bricks Custom AgentsAgent Bricks AI GatewayUnity Catalog
◆ 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://www.databricks.com/blog/databricks-databricks-transforming-sales-experience-using-genai-agents
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

The Field AI Assistant helps sellers retrieve information, distill insights, and automate repetitive manual tasks, with the stated goal of significantly transforming every aspect of the seller journey and improving sa…

What tools did this team use?

Databricks Mosaic AI, Agent Bricks Custom Agents, Agent Bricks AI Gateway, Unity Catalog.

What results were reported?

Seller journey transformation: significantly transform and assist every aspect of the seller journey; Sales velocity: improving sales velocity (source-reported, not independently verified).

How is this sales ops AI workflow structured?

Seller query or seeded prompt → LLM interprets request intent → Driver agent retrieves and processes data → Automated action orchestration → AI Gateway guardrails and monitoring.

Related sales 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 Sales ops cases →