Harriott SmartStay

A fast‑paced hackathon concept blending AI and accessibility, selected as a Top 6 Finalist for its empathy‑driven design and real‑world impact.

ABOUT

Harriott SmartStay is an AI‑powered concept developed during the ASPIRE Hackathon, a fast‑paced, innovation‑focused challenge that brought together cross‑functional teams to solve real‑world problems under tight time constraints.

Our team set out to reimagine how travelers and hospitality teams could use intelligent insights to create more seamless, personalized stay experiences.

MY ROLE

UX Designer

TIMELINE

48 hours

TOOLS

Figma, CoPilot, Claude AI

The Challenge

Hospitality experiences often rely on fragmented data and reactive decision‑making. Guests lack timely, personalized information, while staff must juggle multiple systems to anticipate needs and issues.

The challenge was to design a solution that:

Converts complex data into meaningful insights
Supports proactive decision‑making rather than reactive fixes
Feels useful and human, not overly technical or intrusive.

Team Collaboration

This project was a true team effort, combining perspectives across design, product, and technology. We worked closely to:

Rapidly align on the problem space and success criteria
Brainstorm and prioritize ideas that balanced feasibility and impact
Share feedback continuously to refine the concept under time pressure

Collaboration was constant and iterative, with clear ownership and trust across the team to move quickly while maintaining quality.

My Individual Role

As the Experience Designer on the team, I focused on shaping how the product would feel and function for users. My key contributions included:

Defining the end‑to‑end user experience, mapping how users would interact with the product across key moments of a stay
Designing core AI‑driven user flows and interfaces, ensuring insights were understandable, actionable, and not overwhelming
Translating abstract AI ideas into clear, intuitive prototypes that could be easily communicated to judges and stakeholders
Advocating for clarity and empathy in AI interactions, ensuring the experience explained insights rather than presenting them as black‑box outputs

My goal was to make the AI feel like a helpful assistant—supportive, explainable, and aligned with real user needs.

Design Solution

Harriott SmartStay was designed around AI‑driven insights presented with clarity and context.

Key design decisions included:

Using AI to surface trends, patterns, and recommendations, rather than raw data
Prioritizing insights based on relevance and timing
Designing visual cues and language that clearly explain why something matters
Keeping the interface simple, approachable, and easy to scan

Low‑to‑mid fidelity prototypes were used to quickly test and communicate these ideas, allowing the team to iterate rapidly within the hackathon timeframe.

Features Designed

Within the hackathon timeframe, we designed a focused set of features that demonstrated the core value of the concept. I played a key role in defining and designing these experience elements:

AI Insight Dashboard: A centralized view that surfaces personalized insights about guest behavior, preferences, and potential issues—prioritized by relevance and urgency.

Proactive Recommendations: AI‑generated suggestions presented with clear context, helping users understand what is happening and why it matters.

Personalized Stay Guidance: Lightweight prompts and reminders designed to enhance guest experience at the right moment, without feeling intrusive.

Explainability Cues: UI patterns that clarify the reasoning behind AI insights, reinforcing trust and transparency.

These features were intentionally scoped to demonstrate value quickly while remaining flexible for future expansion.

See Prototype

Future Possibilities

If taken beyond the hackathon, Harriott SmartStay could expand in several directions:

Deeper Personalization: Using longitudinal guest data to tailor experiences over multiple stays.

Predictive Issue Prevention : Surfacing early warnings for potential service issues before they impact guests.

Omnichannel Integration : Extending insights across mobile, in‑room, and staff‑facing tools for continuity.

Learning Feedback Loops : Allowing users to validate or dismiss insights, helping the AI improve over time.

These opportunities reinforce how the concept could scale while maintaining a human‑centered approach.

Reflection

The main challenge was the time crunch and coming up with an idea on how can we best design for people with disability. However, each team member was supportive and we made sure to have regular check-ins and divide work equally.

However, this project reinforced the value of designing with speed, intention, and collaboration. Under time pressure, I learned to prioritize clarity over completeness and to design AI experiences that emphasize trust and usability over novelty. The experience strengthened my ability to collaborate deeply while owning critical UX decisions—translating complex ideas into intuitive, human‑centered experiences.