AI in practice
Client names, sectors, and identifying details remain confidential. The examples below describe real delivery work at the level we are permitted to disclose.
Production system · Anonymous enterprise deployment
Voice AI agent for handling and structuring requests
A production voice agent that conducts adaptive phone conversations, validates the information it collects, and passes structured JSON to the client system.
Challenge
Handling a high volume of requests required consultants to make phone calls, ask follow-up questions, clarify ambiguous answers, and manually convert free-form statements into structured data for downstream processes.
What Inero delivered
- Telephony integration and configuration of ElevenLabs.
- Conversation scenarios and adaptive dialogue logic based on client requirements and historical consultant calls.
- Identification of missing information, contextual follow-up questions, data extraction and validation.
- JSON integration with the client system, security controls, backups, archiving, monitoring, and ongoing helpdesk support.
Confirmed results
- The end-to-end process was reduced from days to hours.
- Manual work and operating costs were reduced while conversations and output data became more consistent.
- The system handles from tens to hundreds of calls per day without requiring the consulting team to scale at the same rate.
- Calls can be made at times convenient for recipients, including between 16:00 and 20:00.
Every conversation and processing step is recorded and archived in line with GDPR requirements and the client’s policies. Missing optional values are returned as null; failure to obtain critical information routes the case for manual review.
Conversational AIVoice agentsElevenLabsTelephonyLLMProduction
Pilot · Anonymous enterprise project
Personalised AI assistant for services and promotions
An AI assistant embedded in a mobile application that helps users find relevant promotions and understand offer conditions using controlled client data.
Challenge
A large and changing set of promotions, regulations, and service variants made it difficult for users to identify offers relevant to their current situation.
What Inero delivered
- The AI layer and conversation scenarios based on client requirements.
- Backend integrations with internal offer data and the client knowledge base, including PDFs and regulations.
- Personalisation using purchase history, services, profile, location, and current offer availability.
- Testing of responses and fallback behaviour when an answer cannot be confirmed in the approved sources.
Pilot objectives
- Increase the use of relevant promotions and the number of purchases made through the application.
- Help users find the right offer and understand its conditions with less effort.
- Encourage use of services on selected days and at selected times to distribute operational load more evenly.
- Verify whether the assistant can resolve common offer questions without involving a consultant.
The assistant does not invent an answer when the source data is insufficient. It states that it does not know or that the requested promotion is not available. The AI layer runs in a cloud environment within the EEA, and the solution is designed in line with GDPR requirements so data processed by the assistant does not leave the EEA.
PersonalisationConversational AIRecommendationsMobileIntegrationsPilot