Customer Service Automation · Enterprise Knowledge Search
Allianz Direct wanted to use generative AI to remove mundane contact-center tasks while keeping customer-service answers trustworthy, compliant and built on the same governed data platform the company already used.
Reported impact
The new agent-assist workflow improved answer accuracy by 10 to 15%, built agent trust in the tool, reduced time spent searching across multiple systems and gave Allianz Direct a governed foundation for rolling out more GenAI use cases across contact centers.
Allianz Direct launched its first generative-AI customer-service workflow on the Databricks Data + AI Platform so contact-center agents could spend less time hunting through policy terms and more time helping customers. The team built a RAG-based agent-assist application that loads Allianz Direct's publicly shareable terms and conditions, retrieves the right passages for each policy question and presents grounded answers to a human agent rather than directly to the caller. Agent Bricks Custom Agents provides the core GenAI capability, Databricks Notebooks make workflow changes easy for developers, and Unity Catalog keeps the broader company data estate securely governed as more teams adopt self-service AI. Because the application runs on the same data and AI foundation Allianz Direct already used, the team could prototype quickly, test with a subset of agents and expand from the initial proof of concept toward broader rollouts and additional GenAI use cases.
Publicly shareable policy terms and conditions are retrieved into a RAG workflow on Databricks, where Agent Bricks Custom Agents generates grounded responses for human contact-center agents. Developers update the workflow in Databricks Notebooks, and governed platform access remains controlled through Unity Catalog as the use case scales.
Compared with the previous application, the Databricks Agent Bricks Custom Agents-powered use case demonstrated a 10 to 15% uplift in accuracy.
Thanks to Databricks, everyone in Allianz Direct has access to the data they need to do their job, securely governed via Unity Catalog.
The team built a proof of concept for a RAG-based agent-assist tool.
This application incorporated all of Allianz Direct products' terms and conditions, intelligently delivering the right information to the agent at the right time.
We don't want shorter calls, we're not trying to replace agents with automation — we're getting rid of the mundane stuff agents don't like doing so they can spend more time talking to customers.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Databricks Sample Aug 30, 2023 Oakbrook Stringent regulation and outdated approaches to consumer lending have made it more difficult for many consumers across the UK to obtain loans. However, fintech organizations such as Oakbrook Finance are revolutionizing… Indeed, the company has been able to launch two new products over the last 12 months, as well as a top-up loan… BigQuery · Databricks Data + AI Platform · Delta Lake Databricks Sample Feb 6, 2023 Ophelos With the UK currently facing the worst cost of living crisis in over 40 years, supporting customers through financial difficulty is quickly becoming one of today’s most pressing challenges. Ophelos is on a mission to… 25% Increase in customers resolving their debts Agent Bricks · Databricks Data + AI Platform · Delta Lake 
Google Cloud Sample Archived · Sep 17, 2026 Home Credit Bank Home Credit Bank needed a safer enterprise AI environment that non-technical teams could use to build banking agents quickly without creating data-leak risk or relying on a long custom-development cycle for each new use… Achieved 100% migration to a secure AI environment Agent Designer · Gemini Enterprise 