product design & data viz

Project . Explainable AI . Dataviz

Move from model explanation to human understanding.

A design exploration of how interaction and visualization can help people inspect, compare, and reason about AI predictions — side-by-side model comparisons, SHAP visualizations, and a structured conversational layer for cohort analysis.

Project . Scoping Review . Data Visualization

Citation Network

An interactive citation graph mapping how closely AI-explanation research and design’s comprehension research actually cite each other — making the pattern something you can see and explore, rather than just state.

Citation network graph showing two clusters of research papers connected by a thin bridge of cross-field anchor nodes
Project . Finance . Internal Tools

Fund Performance Analytics

An internal financial dashboard that brings scattered fund data into one place, balancing quick reference with deeper performance and comparative analysis.

NAV Historical chart showing fund price trends with numbered callout pins, date controls, and the quartile legend
Project . Music Recommendation . Thesis

Songflow

A master’s thesis exploring how explanations can become interactive—helping people understand music recommendations and use descriptive attributes to refine them.

Songflow prototype showing tag-filtered recommendations for White Winter Hymnal by Fleet Foxes, with tags like folk, indie, and mellow linked to recommended tracks
Project . Data Visualization

How Oscar Nominations Connect

Explore the hidden relationships between Academy Award categories.

Built with Claude Code and Cursor.

Chord diagrams showing how Oscar nomination categories connect — Best Picture, Directing, Writing, and acting categories linked by films nominated across multiple categories
Project . Data Visualization

Is LEGO Going Gray?

Is LEGO going gray? How color palettes shift as more Lego sets target adults.

Built with GitHub Copilot and Cursor.

Ten pie charts showing the color palette breakdown of LEGO sets by theme — Friends, Star Wars, Harry Potter Universe, Technic, Disney, Ultimate Collector Series, Batman, Botanicals, Duplo, and Architecture

Hi, I'm Juliana.

I'm a product and data visualization designer specializing in UX for power users working with complex, high-volume data—designing dashboards, analytics tools, and decision-support systems grounded in research and analysis.

Clients include Goldman Sachs, JPMorgan Chase, Morgan Stanley, and ADP.

This site is a space for personal explorations, where I treat code as a design medium, using AI-assisted tools to rapidly explore interaction patterns, data representations, and system behaviors—particularly in the context of explainable AI and model-informed decision-making.

To get in touch, email me at bigpixel@gmail.com.