Fish Eyes

AI-Assisted Museum Analytics Tool

Role: UX Designer / Team Lead
Tools: Figma, User Research, Prototyping, Presentation Design
Context: NYC Media Lab Combine III + Verizon Connected Futures
Team: 3–4 designers / technologists

Overview

Fish Eyes (formerly CurAItor) is an AI-assisted analytics tool designed to help museums understand visitor behavior using existing security camera footage. The system analyzes audience movement patterns to surface insights about traffic flow, engagement zones, and underutilized spaces, without requiring new hardware or invasive tracking methods.

This project focused on translating complex machine-learning outputs into readable, actionable insights for non-technical museum staff.

Problem

Museums face declining attendance and engagement but lack accessible tools to understand how visitors actually move through exhibits.

Common constraints included:

• Limited budgets for new tracking infrastructure

• Staff without technical or data analysis backgrounds

• Existing footage that was underutilized beyond security purposes

The challenge was not data collection, but rather making behavioral data legible and useful.

Design Question

"How might we present AI-generated visitor data in a way that museum staff can interpret, trust, and act on without needing technical expertise?"

Solution

Fish Eyes uses computer vision to process existing video footage and generates visual overlays and summaries that highlight:

• High-traffic vs low-engagement zones

• Dwell time around exhibits

• Bottlenecks and circulation issues

The UX prioritizes:

• Visual-first insights over raw metrics

• Comparisons across time (weekday vs weekend, special events, etc.)

• Clear affordances for non-technical users

Rather than asking staff to “analyze data,” the interface answers practical questions like:

• Which exhibits attract the most attention?

• Where do visitors tend to stop or avoid?

• How does layout affect flow during peak hours?

Clickable prototype

My Role & Contributions

As team lead, I was responsible for:

• Defining the core UX problem and product scope

• Delegating responsibilities across design and technical team members

• Conducting exploratory research through cold outreach to museum staff

• Translating stakeholder pain points into product requirements

• Leading presentations for NYC Media Lab and Verizon stakeholders

I also contributed directly to:

• Interaction flows and interface hierarchy

• Data visualization concepts

• Narrative framing for demos and pitches

Design Considerations & Tradeoffs

• Privacy vs insight: Designed for anonymized pattern analysis rather than individual tracking

• Accuracy vs interpretability: Prioritized readable trends over exposing raw AI confidence scores

• Flexibility vs simplicity: Limited customization to avoid overwhelming staff with options

These tradeoffs were intentional to keep the tool usable within real institutional constraints.

Outcome

Fish Eyes was:

• Selected for NYC Media Lab Combine III

• Featured in Verizon Connected Futures

• Awarded “Best in Show” at an AI/UI Design Jam

The project demonstrated how emerging technologies can be grounded in practical UX design to support real-world decision-making.

Branding poster

Why This Project Still Matters

This work reflects my ongoing approach to UX and systems design:

• Designing for non-expert users

• Making complex systems legible

• Balancing ambition with institutional constraints

• Leading multidisciplinary teams through ambiguity

You may also like