Define a stronger AI visual language.
I focused on how AI-specific cards, charts, and metrics should feel more unified across the product.
Project / 03
AI Pattern and Visual Language Unification
A ServiceNow internship project focused on unifying the AI visual language inside AI Control Tower so governance, inventory, and value signals feel clearer, more recognizable, and more scalable.
AI Control Tower is a centralized dashboard for monitoring AI usage, governance, and performance across enterprise systems. My role was to make the product feel more consistent by building a stronger AI-specific visual language across its most important surfaces.
The work focused less on adding new features and more on sharpening recognition, improving hierarchy, and making dense AI oversight information easier to parse at a glance.
Discuss this project ↗Make oversight information easier to read fast.
The priority was better recognition, clearer hierarchy, and more consistent behavior across dense product surfaces.
Dashboard, inventory, governance, and value refinements.
The work translated directly into visual improvements across key AI Control Tower modules rather than a speculative concept only.
Challenge
The interface worked, but the AI identity felt fragmented.
Different cards, charts, and data moments looked like they belonged to different systems. That inconsistency weakened recognition and made the experience feel less cohesive than the importance of the product demanded.
Fragmented visual language.
Cards and data modules were inconsistent in spacing, chart treatment, and emphasis, which made AI-specific information feel visually disconnected.
No clear AI identity.
The system needed recognizable cues that could distinguish AI signals from standard enterprise dashboard content without becoming noisy or decorative.
Hierarchy was not doing enough.
Key metrics like value, posture, and risk needed stronger visual prioritization so users could understand what mattered first.
System Moves
The work translated broad principles into repeatable visual decisions.
Rather than treating each card as a one-off redesign, I used the project to establish reusable moves for chart emphasis, metric hierarchy, and AI-specific recognition.
Make key metrics speak sooner.
Value, posture, and risk cards were adjusted so headline information landed first, with supporting patterns and trend detail structured underneath.
Let charts feel clearer without feeling heavier.
Graphs and data moments became easier to parse through more deliberate contrast, cleaner trend shapes, and reduced visual ambiguity.
Build cues users can learn once and reuse.
The resulting AI language was meant to scale, so the same logic could carry across multiple surfaces without re-explaining itself.
Principles
Consistency, recognition, scalability, and clarity guided every refinement.
I used these principles to decide how charts should read, how AI cards should feel, and how the overall system could scale across multiple AI Control Tower surfaces.
Make every module feel related.
Shared treatment across cards, data visualization, and layout created a more unified product rhythm.
Give AI content its own presence.
Subtle AI-specific visual cues improved immediate recognition without breaking the platform's enterprise tone.
Refinements had to work across many surfaces.
The system was designed to hold up across dashboard, inventory, governance, and future AI oversight modules.
Deliverables
Refining value, governance, and risk cards into a more coherent AI language.
The project translated principles into direct interface changes: cleaner charts, stronger emphasis, clearer state differences, and more consistent visual behavior across dashboard components.
Reflection
Small interface shifts created a stronger sense of trust and product identity.
The final learning from this project was that consistency is not just polish. In AI products, consistency helps users trust what they are seeing, and visual language becomes part of how the product communicates intelligence and reliability.
Consistency builds trust.
When data, charts, and cards align visually, the product feels more dependable and easier to interpret.
Visual language signals product identity.
A clearer AI-specific system helped AI Control Tower feel more intentional and recognizable as its own experience.
Small changes can reshape the whole surface.
Subtle hierarchy, chart, and color refinements made the interface feel more coherent without requiring a full redesign.
