Customer Service Automation · Enterprise Knowledge Search
DI Square needed to improve call-center operator productivity and reduce dependence on individual know-how as labor shortages and higher staffing costs made it harder to hire, train, and retain operators.
Reported impact
DI Square created a call-center knowledge assistant that helps operators search faster, standardize answers, and increase response productivity while preserving and reusing accumulated know-how.
DI Square extended its EasyAnswer Series with EasyAnswer JINN/RAG-Opt, a call-center support capability built around a custom retrieval-augmented-generation architecture using IBM AI services. The system links company data such as customer information, inquiry history, manuals, regulations, and product data to IBM watsonx Assistant, IBM Watson Discovery, and the Japanese version of IBM Granite delivered through watsonx.ai. Operators type questions into the interface and the assistant searches company knowledge to present appropriate answer candidates and relevant past conversations, helping them respond faster and more consistently. Over time, usage feeds new findings and corrected knowledge back into the system so answer quality can improve and standardize across the operation. In practice, the solution becomes a governed knowledge-assistance layer for call centers, combining enterprise retrieval with Japanese-language LLM performance tuned for business use.
Customer information, inquiry history, manuals, regulations, and product data are indexed into the RAG architecture, operators submit questions through watsonx Assistant, Watson Discovery retrieves relevant knowledge, and Granite on watsonx.ai generates answer candidates and uses past conversation history to support the response.
"EasyAnswer JINN/RAG-Opt" embeds watsonx Assistant, Watson Discovery and the Japanese version of Granite provided by watsonx.ai.
"EasyAnswer JINN/RAG-Opt", with its unique RAG architecture, enables search results based on the company's data
This function helps operators to do efficient searches and to use past conversations history to instantly give effective responses to their customers.
helping to resolve various issues these are challenged with
the quality of answers provided by the AI can be improved and standardized
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
ServiceNow Sample Modified · Jul 30, 2026 CANCOM CANCOM had siloed data across ERP, CRM, SharePoint, and other systems, which made it hard to streamline services, reuse enterprise knowledge, and give employees or customers consistent support. 1 Unified platform supporting all internal services Now Assist · ServiceNow AI Platform · ServiceNow portal ServiceNow Sample Modified · Jul 30, 2026 LAB3 Pty Ltd LAB3 needed to modernize service management so support engineers were not spending a week chasing follow-ups, validating request items manually, and answering routine queries instead of delivering higher-value customer… 100x increase in usage via the service portal AI Search · Agent Workspace · Creator 
Google Cloud Sample Archived · Jul 25, 2026 Proforce Pest Control Headquartered in Boca Raton, Florida, Proforce Pest Control provides residential pest control across 13 branches in five Southeastern and Mid-Atlantic states. The business grew out of a series of acquisitions,… 22% less fuel per appointment after route optimizations BigQuery · Cloud Run · Gemini 
