SUCCESS CASE

Platform of
AI agents
for conversation analysis and support
in the automotive sector

We developed a multi-agent platform on Azure AI for a Korean automaker, automating the transcription and analysis of 500 daily contact center audios. The solution generated sentiment and quality insights at lower costs than traditional platforms.

Azure OpenAI
Multimodal AI
GenAI

Results

customer service audios processed daily with high-precision transcription
0

Reduced operational and licensing costs compared to traditional platforms

Continuous generation of insights on quality, compliance, and customer sentiment

Processing with real-time feedback for tracking the analyses

Overview

A Korean automaker was facing a bottleneck in its operations when handling a high daily volume of audio interactions coming from dealerships and contact centers. With manual and inconsistent analysis, the organization had low analytical coverage and difficulties monitoring service quality, conversation compliance, and customer sentiment, preventing the strategic use of this data.

The challenges

Manual, limited, and inconsistent analysis of a large daily volume of audio interactions.
Low visibility into service quality and operational compliance.
Difficulty extracting customer sentiment analysis and converting it into decisions.
High costs associated with using traditional corporate platforms on the market.

Journey
and solutions

We developed a platform with a multi-agent architecture using LangChain, LangGraph, and Azure AI. The sequential flow transcribes audio with high precision via Whisper/FFmpeg, separates speakers, and analyzes emotional tone and sentiment. The solution generates automated executive summaries and offers real-time tracking of each step via Server-Sent Events (SSE) within a scalable enterprise structure.

What did we deliver?

Specialized multi-agent architecture built with LangChain, LangGraph, and Azure AI.
High-precision audio-to-text transcription module using Whisper and FFmpeg.
Intelligent conversation analysis engine with diarization, tone, and sentiment.
Enterprise architecture with support for multiple formats and real-time notifications.

More success stories

Financial Services Cloud on one of the largest digital platforms in the world

ília implemented Salesforce Financial Services Cloud and CRM Analytics for a financial services giant. The solution centralized the 360° view of portfolios, increasing advisor productivity to serve R$ 30 billion in assets with scale and retention.

Interseguro: Product Discovery for the digital renewal of SOAT insurance

ília conducted the Product Discovery for the Peruvian insurance company Interseguro, restructuring the web and mobile experience of SOAT insurance. Validation with high-fidelity prototypes and usability testing simplified journeys and increased the adoption of the digital product.

24/7 Customer Service with AI and WhatsApp: Scaling with Governance in the Insurance Sector

We structured a solution with Salesforce Einstein Bot and WhatsApp for a large insurance and pension company. The automation ensured 24/7 self-service, reducing the load on the human customer support team and unifying the customer journey with regulatory compliance.

Data Lake and AI: Operational efficiency and predictive power for energy generation

We centralized data and applied AI for a multinational energy company. With a Data Lake in medallion architecture and predictive models, the solution enabled predicting equipment failures, speeding up decision-making, and moving from reactive to predictive management.