PROJECT OUTCOMES
Overall user satisfaction
Reduced coaching feedback delays by
Cut down feedback navigation time by
OVERVIEW
CLIENT REQUIREMENT
To build a system that leverages AI with human expertise to analyze audio recordings of teaching sessions, provide personalized feedback, and align insights with existing teaching frameworks.
My Role
Product Design
Strategy Design
Domain
Edutech
AI
THE CHALLENGE
The project started with ambiguity. Our client, envisioned an AI-powered coaching tool—
Vision
Audio Record Teaching Sessions
│
AI Analyzes Audio Against Teaching Standards
│
AI Generates Feedback Suggestions
│
Human coach Reviews & Adjusts Feedback
│
Actionable Improvements for Teaching
But, the scope, technical feasibility, and user needs were unclear.
THE CHALLENGE
Feedback delays exceeding 72 hours.
Generic, one-size-fits-all guidance.
Teachers struggling to manage time while providing quality insights.
Limited access to AI tools that could assist without adding cognitive load.
MY ROLE & APPROACH
As the Founding Product Designer, I owned the feedback page end-to-end—from research through prototyping and testing. My responsibilities included:
Conducting user research: 5 teacher interviews, 12 surveys, 20+ research papers, 7 competitor analyses.
Synthesizing insights into personas, user stories, and use cases.
Designing AI-assisted feedback flows with clarity, usability, and scalability in mind.
Building interactive prototypes and conducting iterative usability testing.
Collaborating with 4 designers, a PhD client, and subject matter experts to ensure pedagogical accuracy and technical feasibility.
THE PROCESS
Research & Empathy:
We conducted research combining surveys, interviews, and competitive analysis to understand teachers’ needs and the role of AI in education.
Figure shows: Data collection and Analysis on Figjam.
67%
Teachers would use AI-assisted feedback.
75%
Teachers are comfortable being recorded in class.
86%
Teachers believe technology should be part of the classroom.
40%
Currently have coaching support; feedback is often delayed or verbal.
Empathy mapping revealed teachers want clear, bite-sized guidance, privacy-respecting observation, and tools that save time.
Figure shows: Data collection and Analysis on Figjam.
Competitive analysis highlighted a gap in AI integration, presenting an opportunity for personalized, standards-aligned feedback.
Figure shows: Data collection and Analysis on Figjam.
THE PROCESS
Empathy & Research:
Defining the Problem:
Synthesizing research led to clear personas and user journeys. The primary insight: teachers wanted AI to assist, not replace, and for feedback to be actionable, personalized, and trackable.
Design & Prototyping:
I led iterative design on the feedback page:
Before: Long, unstructured scrolling; slow navigation.
After: Time-stamped, AI-assisted dynamic chat that highlighted key insights.
Prototypes were refined through multiple rounds, each incorporating teacher feedback.
Testing & Validation:
Usability tests revealed:
Navigation time reduced by 40% (from 6 min → 3.5 min).
90% of teachers reported satisfaction with clarity and speed.
Feedback delays reduced 100%, enabling near real-time student guidance.
Bridging Human + AI:
To ensure adoption, I designed interactions that made AI suggestions transparent, editable, and context-aware—so teachers retained control while gaining efficiency.
THE OUTCOME
The final prototype delivered measurable impact:
Faster, actionable feedback for students.
Increased teacher efficiency, saving hours per week.
Validated AI-human collaboration, ensuring trust and adoption.
The project showcased my ability to:
Lead end-to-end design in ambiguous, AI-driven environments.
Turn complex workflows into intuitive, elegant interfaces.
Collaborate cross-functionally with researchers, designers, and engineers.
Use metrics to validate design decisions.
REFLECTION
This project reinforced the power of empathy-driven, AI-assisted design. Key learnings:
Rapid iteration is critical when designing emerging tech workflows.
Clear communication and research synthesis accelerates cross-functional alignment.
Balancing automation with human agency ensures adoption and trust.
If scaled to production, the approach could transform teacher-student interactions, reducing friction, increasing engagement, and shaping the future of AI in education.
