First place — CCC mock case
The challenge: reduce friction in travel planning and make the portal feel more relevant to each traveler. Our team proposed tailoring the experience around customer needs, preferences, and context.
My independent AI/ML contribution
I independently owned the AI and machine learning recommendation component of our team’s proposal. I worked on the recommended technical approach for user segmentation and personalized travel suggestions, including how recommendation logic could feed into a conversational travel assistant. The wider consulting narrative, competitive framing, and implementation plan were team deliverables.
This was a proposed AI/ML design and concept, not a claim that a production recommendation engine was trained or launched.
How the proposed system fits together
Use relevant customer signals such as travel preferences, history, card type, and spending patterns, subject to appropriate data permissions.
Explore dynamic customer clustering to tailor destination, accommodation, and activity suggestions for differing travel styles.
Surface personalized options in an AI travel assistant, with feedback, sentiment analysis, and fraud/risk safeguards as proposed capabilities.
From idea to business plan
The broader mock-case presentation included customer personas, example chatbot interactions, an estimated multiyear cost model, risk mitigations, and a staged implementation roadmap through 2030. These were planning scenarios rather than measured deployment outcomes.
The takeaway
A useful ML proposal needs more than a model idea. It needs to fit the user experience, work with existing business incentives, address responsible data use, and have a credible path from prototype to evaluation.
Explore the work
Read the original project materials behind this case study.
Student mock-case competition submission; not an official Capital One engagement or a deployed product.
AI Music Generation