
Sana Coach
Turning fragmented talent workflows into a continuous AI coaching model.
AI Product Strategy · Design Engineering · Systems Thinking
I helped define the interaction model for an AI-first coaching experience connecting performance, feedback, learning, skills, and career development—and built the React prototype used by Product, Engineering, Design, and leadership to evaluate the direction.
AT A GLANCE
ROLE
Senior UX Design Engineer
TOOLS
Figma · React · TypeScript · GitHub · Workday Canvas · Sana Design System
TIMELINE
June 2026 – November 2026
PARTNERS
Product · Engineering · Design · Leadership
SCOPE
Performance Management · Feedback · Learning · Skills · Career Development
WHAT I OWNED
Interaction model · AI behavior · Connected workflows · High-fidelity React prototyping · Reusable patterns
CHALLENGE
Performance management was organized around individual events—goals, feedback, reviews, development conversations, learning, and career planning—but the context connecting those moments was often lost.
Managers repeatedly had to reconstruct what had happened, understand progress, gather feedback, identify development opportunities, and decide what to do next. Employees faced the same fragmentation when trying to understand their growth and direction.
The bigger question was not simply how to make each workflow faster. It was how an AI system could maintain context over time and help continuously, rather than waiting for someone to initiate another HR task.
STRATEGIC SHIFT
The work required us to stop thinking about performance management as a collection of HR workflows and start thinking about it as an ongoing relationship between the employee, manager, AI, and the broader talent system.
That changed the design problem. Instead of asking only "How do we make this task easier?" we also had to ask the following:
What does the system know?
When should it intervene?
What context should carry forward?
What should AI recommend versus act on?
When should it stay out of the way?
Where must human judgment remain in control?
The central principle of the work then became:
AI product design is behavioral design.
The interface matters, but so does defining how the system behaves over time.
MY ROLE
Context is a product capability
The usefulness of an intelligent system increases dramatically when it can connect information across time and across workflows.
Designing that continuity is as important as designing the individual screens.
Simplicity often sits on top of complexity
The best experience may look conversational and lightweight while relying on a sophisticated system underneath.
A large part of my role was absorbing that complexity and exposing only what the user actually needed.
KEY DECISIONS
Decision 1 — When should AI intervene?
The tension was between usefulness and interruption. If the system waited for every explicit request, it behaved like another tool. If it acted too aggressively, it risked feeling intrusive or opaque.
I explored a behavioral model in which the system could assist, recommend, prompt, or stay out of the way depending on context and consequence. The goal was to make AI feel proactive without making it presumptive.
Decision 2 — How much context should carry forward?
Performance, feedback, learning, skills, and career development were traditionally separate workflows. Treating each interaction as a fresh start would preserve that fragmentation even inside an AI experience.
I designed around continuity: relevant context should accumulate over time so the system can help users without repeatedly asking them to reconstruct the same story.
Decision 3 — Where does human judgment remain essential?
Many performance decisions affect careers, compensation, growth, and trust. The system could surface context and recommendations, but it should not obscure who is responsible for the final decision.
The interaction model therefore emphasized assistance and explanation while keeping consequential judgment with the manager or employee.
HOW I WORKED
A shared product model
Teams had a clearer model for how AI could connect previously fragmented talent experiences.
Context became part of the experience
Information could carry across moments instead of each workflow starting from zero.
AI could help proactively
The experience wasn't limited to user-initiated tasks.
Human judgment remained explicit
AI could inform consequential decisions without obscuring who remained responsible for them.
Separate workflows began to feel connected
Performance, feedback, development, learning, and coaching could behave as parts of the same system.
WHAT CHANGED
A shared product model
Teams had a clearer model for how AI could connect previously fragmented talent experiences.
Context became part of the experience
Information could carry across moments instead of each workflow starting from zero.
AI could help proactively
The experience wasn't limited to user-initiated tasks.
Human judgment remained explicit
AI could inform consequential decisions without obscuring who remained responsible for them.
Separate workflows began to feel connected
Performance, feedback, development, learning, and coaching could behave as parts of the same system.
LEARNINGS
Context is a product capability
The usefulness of an intelligent system increases dramatically when it can connect information across time and across workflows.
Designing that continuity is as important as designing the individual screens.
Simplicity often sits on top of complexity
The best experience may look conversational and lightweight while relying on a sophisticated system underneath.
A large part of my role was absorbing that complexity and exposing only what the user actually needed.
CONTRIBUTIONS
Product Design
AI-first interaction models · Continuous performance management · Coaching experiences · Connected talent workflows
Design Engineering
React prototyping · TypeScript · GitHub collaboration · Design-system implementation · Reusable components
Systems Thinking
Performance Management · Feedback · Learning · Skills · Career development · Talent mobility
Leadership
Cross-functional alignment · Product exploration · Prototyping strategy · Mentorship · Design-engineering practice