Steward
(knowledge transfer tool)
Overview
Summary
Steward is an end-to-end Knowledge Transfer and automation platform designed to streamline standard operating procedure generation, data analysis, and workflow transition. The platform enables organizations to capture institutional knowledge and deliver it through structured learning paths.
Project Scope
The project focused on designing intuitive interfaces for knowledge transfer workflows, AI-assisted content generation, and dashboard visualizations that help users navigate complex organizational data.
My Role
Product & Interaction Design
I was part of UX design team for Steward's core workflows, collaborating with project managers to translate complex knowledge transfer requirements into intuitive, scalable interfaces.
- Dashboard UI & Visualization
- AI Canvas Interaction
- Cross-App Research Mapping
Specific Area Of Focus
Dashboard UI Design & Data Visualization
Redesigned the dashboard to surface key metrics and workflow status at a glance, balancing information density with clarity for enterprise users.
- Restructuring the dashboard layout for improved scanability.
- Designing data visualization components for workflow metrics.
- Creating responsive layouts for desktop.
Agentic AI Canvas UI & Interaction Design
Designed the Agentic AI Canvas.
A workspace where users interact with AI agents to generate, refine, and organize knowledge transfer content.
- Agentic AI Canvas layout with main workspace and AI cat interface
- Conversational UX With AI that displays video/document transcriptions, and learning paths.
Information Architecture Mapping
Steward was created by combining two existing applications (Klewer and Quasar). This foundational knowledge allowed me to design the UI components according to the new system's navigational structure
Value and Impact
- AI-driven solutions that accelerate knowledge transfer workflows
- Reduced Cognitive Load through structured information hierarchy
- Improved onboarding efficiency for new team members
- Scalable design patterns for future platform features
