Customer Service Automation · Enterprise Knowledge Search
INAIL needed to absorb a surge in grant-funding inquiries and application volume for its Bandi ISI program without overwhelming the contact center or delaying applicant guidance during the submission and admission process.
Reported impact
INAIL improved the consistency and availability of applicant support, eased pressure on its contact center during a high-demand funding round, and created a knowledge base it can reuse for future review and verification workflows.
INAIL worked with IBM Consulting and IBM Client Engineering to build a virtual assistant for its Bandi ISI funding process. The pilot used unstructured content such as manuals, FAQs, prior tickets, requests, and official documentation to power a conversational support layer embedded in the web application companies use to access the funding procedure. IBM watsonx Assistant acts as the orchestration layer between IBM Watson Discovery and IBM watsonx.ai: Discovery indexes the documentation and identifies the text passages likely to contain answers, while watsonx.ai runs the Mixtral 8x7B Mistral model to generate response text grounded in those indexed materials. This architecture lets applicants ask both general and specific questions about the call, eligibility, and operating procedures, while INAIL can standardize contact-center responses and later extend the same knowledge base into automated verification during post-approval review phases.
Applicants submit natural-language questions in the Bandi ISI web application, watsonx Assistant orchestrates the request, Watson Discovery retrieves relevant passages from manuals and prior support content, and the Mistral model on watsonx.ai generates the final response for applicants or contact-center staff.
24x7 assistance, eliminating wait times and the risk of missed deadlines.
50% rise in applications
Finally, the gen AI Mistral LLM model (mixtral8-7B) on watsonx.ai was used to engage the conversation and generate answer texts
IBM Watson Discovery was used as a documentation collector and cognitive engine available to the overall application.
The solution involved IBM watsonx Assistant, which was able to understand and address various questions, general or specific, by leveraging multiple features and acting as the orchestrator between IBM Watson Discovery platform and IBM watsonx.ai AI studio.
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
IBM Sample Apr 2, 2024 South Waikato District Council South Waikato District Council needed a better way for citizens and employees to find information hidden across siloed public-service repositories because its first attempt at a virtual assistant lacked the technical… 91.5% accuracy rate validated across 300 test questions IBM Watson Discovery · IBM watsonx Assistant · IBM watsonx.ai 
Google Cloud Sample Archived · Oct 2, 2026 Hawaii DCCA The Hawaii Department of Commerce and Consumer Affairs (DCCA) serves a vital mission: protecting residents while ensuring a level playing field that promotes fair business competition across the state. Reclaiming Time: Streamlined operations and reduced manual work across 52 state boards and commissions by… Cloud Storage · Contact Center AI Platform · Customer Experience Agent Studio 
Baidu AI Cloud Sample Sep 24, 2025 All-China Federation of Trade Unions The All-China Federation of Trade Unions needed a unified AI foundation that could support legal aid, rights protection, and other worker services at national scale instead of leaving union service digitization… Delivery cycle fell from 3-6 months to under 1 month. Qianfan AppBuilder · Qianfan ModelBuilder · Union Industry Model 