case study

Car IQ

Unleashing Operational Efficiency at Car IQ with Google Cloud LLM-driven Intelligent AI Assistant

Business Impacts

Enhanced Efficiency: GenAI Assistant streamlines CarIQ's operations, saving time and resources for increased productivity

Knowledge Compliance: The RAG framework ensures accurate policy information, promoting accessibility and strict adherence to company policies for Car IQ

Customer Key Facts

  • Country : USA
  • Industry : Fintech
  • Website : www.cariqpay.com

Problem Context

Car IQ, a fintech startup, specializes in streamlining cardless transactions at fuel stations for drivers associated with fleet management companies.

Car IQ’s product team expressed a keen interest in exploring Google Cloud’s Generative AI capabilities for the purpose of crafting an Intelligent chatbot. The primary objective was to enable the chatbot to effectively query an internal knowledge base.

Challenges

  • Diverse data sources with varying formats, including pdf, ppt, docs, posed a challenge in extracting accurate and meaningful information
  • Significant experimentation was undertaken to determine the optimal hyperparameter configurations for the Language Model (LLM) to effectively handle the Car IQ dataset
  • The absence of a test dataset, coupled with the subjective nature of LLM-generated answers, made result evaluation challenging

Technologies Used

Vertex AI Workbench

Vertex AI Workbench

Firestore

Firestore

Google Cloud Storage

Google Cloud Storage

Cloud Run

Cloud Run

Artifact Registry

Artifact Registry

Vertex Model Garden

Vertex Model Garden

Cloud Build

Cloud Build

Load Balancer

Load Balancer

GCP IAM

GCP IAM

VPC

VPC

Vertex AI Pipelines

Vertex AI Pipelines

Solution

  • Quantiphi delivered a fully operational chatbot, leveraging Car IQ's knowledge base - facilitating the retrieval of answers to user queries
  • A UI with feedback feature was seamlessly integrated into the chatbot, empowering Car IQ to provide input on the chatbot's responses
  • These feedback entries were automatically stored in a database, contributing to ongoing enhancements in solution performance
  • The solution's automation was achieved through Vertex AI pipelines, allowing the Car IQ team to efficiently update the knowledge base in under 10 minutes by creating a new pipeline run, ensuring swift and effective knowledge base updates

Results

Car IQ benefited from the creation of a reliable chatbot UI using Google Cloud LLM models, facilitating seamless access to answers from their internal knowledge base

"Car IQ benefited a lot from Quantiphi's expertise in delivering Gen AI and Predictive Analytics solutions. They helped accelerate our timelines to market and understand industry best practices. The Quantiphi team was highly proficient, flexible and professional which made working with them a seamless experience for us."

Khalid El-Awady, Chief Product Officer, Car IQ

Thank you for reaching out to us!

Our experts will be in touch with you shortly.

In the meantime, explore our insightful blogs and case studies.

Something went wrong!

Please try it again.

Share