Nissan Dealer Training Analytics & AI Chatbot
RAG Architecture · Amazon QuickSight · Data Analysis · Survey Design
Team Lead | Vanderbilt Capstone with Nissan North America | Aug 2025 - Dec 2025
Led a 3-person team on a capstone project in collaboration with Nissan North America. Our goal was to diagnose how dealers engage with training resources, identify barriers to effective learning, and prototype an AI-powered solution to address their needs.
The project culminated in a presentation to Nissan NA executives, with insights informing ongoing development with an external partner.
The Challenge
Nissan operates a training platform serving thousands of dealers across North America. Despite significant engagement, there was no systematic analysis of how dealers actually use the platform or where they face friction. The question: how can training become a resource dealers actively want to use, rather than just a compliance requirement?
My Approach
Phase 1: Diagnosis
- Qualitative Research: Conducted on-site dealership visits in Nashville and interviewed sales staff and product specialists
- Quantitative Analysis: Analyzed platform engagement logs at scale to identify usage patterns across different job roles
- Survey Design: Deployed a survey directly on the training platform to gauge dealer preferences and interest in AI-powered tools
Phase 2: Solution Prototyping
- Based on research insights, built a proof-of-concept RAG-based AI chatbot focused on the 2026 Nissan Leaf
- Integrated product documentation, web-scraped specifications, and competitor data into a unified knowledge base
Technical Solution
The chatbot was built using a RAG (Retrieval-Augmented Generation) architecture, which grounds LLM responses in a curated knowledge base rather than relying on the model's training data alone.
Architecture
RAG pipeline with Amazon QuickSight agent capabilities
Knowledge Base
Official docs, web-scraped specs, competitor data
Capabilities
Product Q&A, competitor comparisons, mobile-friendly
Data Sources
Nissan documentation, manufacturer sites, forums
Impact
- Presented findings and prototype to Nissan NA executive leadership
- Research insights are informing ongoing chatbot development with an external partner for potential dealer rollout
- Demonstrated clear dealer interest in AI-powered product knowledge tools
What I Contributed
- Led the 3-person team through research, analysis, and prototyping phases
- Built the RAG-based chatbot prototype using Amazon QuickSight
- Designed and deployed the dealer interest survey on the training platform
- Presented findings and demonstrated the prototype to Nissan NA executives