Back office ops · Production

Accolade unifies fragmented healthcare data on Databricks Mosaic AI to enable RAG-powered internal inquiry system

The problem

Accolade's healthcare data was siloed across multiple platforms with no real-time access, hindering accurate member stratification, timely care delivery, and the development of AI-driven initiatives.

Workflow diagram · grounded in source
1
Lakehouse data consolidation
integration
“a lakehouse architecture that enabled the combination of data storage and management to facilitate easier access and analysis”
2
Streaming data ingestion
integration
“the Databricks Platform tapped into Apache Spark™ for streaming data capabilities, enabling continuous data ingestion from various sources”
3
HIPAA-compliant data governance
validation
“Databricks Unity Catalog was critical for management and governance, supporting HIPAA compliance requirements with stringent access controls and detailed data lineage”
4
RAG solution development
ai_action
“Accolade used the Agent Bricks Custom Agents to develop a RAG solution specifically tailored to improve the efficiency and effectiveness of their internal teams. The Databricks Platform allowed the company's power users to access diverse…”
5
DBRX LLM query processing
ai_action
“Accolade's RAG solution relied on DBRX, an open source LLM developed by Databricks, to help them use proprietary enterprise data for more accurate and context-aware outputs. For example, Accolade leveraged DBRX to enhance internal search…”
6
Real-time model serving
output
“Databricks Model Serving came into play. It deployed the model as a RESTful API, enabling real-time predictions that could be integrated directly into Accolade's decision systems”
Reported outcome

Accolade's time to insight greatly improved, with internal teams achieving major productivity gains and the confidence to handle complex member inquiries using a RAG system built on Databricks Mosaic AI.

Reported metrics
Time to insightgreatly improved
Team productivitymajor productivity gains
Reported stack
Databricks Data Intelligence PlatformApache SparkDatabricks Unity CatalogDatabricks Mosaic AIAgent Bricks Custom AgentsDBRXDatabricks Model Serving
Source
https://www.databricks.com/customers/accolade
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Accolade's time to insight greatly improved, with internal teams achieving major productivity gains and the confidence to handle complex member inquiries using a RAG system built on Databricks Mosaic AI.

What tools did this team use?

Databricks Data Intelligence Platform, Apache Spark, Databricks Unity Catalog, Databricks Mosaic AI, Agent Bricks Custom Agents, DBRX, Databricks Model Serving.

What results were reported?

Time to insight: greatly improved; Team productivity: major productivity gains (source-reported, not independently verified).

How is this back office ops AI workflow structured?

Lakehouse data consolidation → Streaming data ingestion → HIPAA-compliant data governance → RAG solution development → DBRX LLM query processing → Real-time model serving.

Related back office ops cases