herupu ai

The problem

Many students struggle to concentrate because of anxiety, distractions and poor study habits. The support they receive (psychologist, professor, guidance department) is uncoordinated: each system works in parallel without connecting the information.

We built Herupu AI to connect all of that support into a single empathetic tool that the student can use every day.

Research

We researched the context and built a persona based on real profiles of students with anxiety:

Thomas, 16, London. He has an anxiety disorder. He does not consider himself a bad student but, no matter how much maths he studies, his results don’t follow. After several failures he lost his confidence, which deepened the anxiety and made concentration even harder.

From Thomas we mapped:

• An 8-point journey map from getting home from class to understanding the material.
• A dramatic arc tracing the emotional level at each point (from frustration to optimism).
• What if?: scenarios where the system fails. What if the psychological support is not effective? What if the device doesn’t explain a concept well? What if the departments don’t talk to each other?
• A service blueprint showing how the departments (guidance, mental health, professors) would connect to Herupu.

Benchmarking

We analyzed four references and positioned Herupu along level of engagement versus study organization:

• Duolingo: high engagement, focused on languages.
• Daylio: daily emotional tracking.
• My Study Life: pure organization, low on emotion.
• Shiken AI: mindfulness plus quizzes.

Herupu sets itself apart by integrating all three axes (organization, emotion, AI tutor) into a single product.

Design process

We followed Design Thinking through these phases:

1. Empathize: interviews, the Thomas persona, journey map.
2. Define: synthesis of the problem, “what if” for risks.
3. Ideate: low-fidelity paper prototyping.
4. Prototype: wireflow → design system → high-fidelity wireframes in Figma.
5. Test: prototype in Maze with real users.

My specific contribution was [spell out here: research, UI, the Figma prototype, whatever it was, write what you did most]. Joana handled [her part] and Ochai [his part].

Outcome

We tested the prototype in Maze with real users in the target range. The data:

• 88% said the app was easy to use.
• 83% felt it recognized their needs.

Those numbers are what counts in a portfolio: evidence that the design works in the hands of someone who isn’t me.

What I learned

• Designing for anxiety demands a level of empathy that no commercial brief asks for. Every piece of microcopy matters.
• Working in a team of 3 with blurred authorship is exactly the real situation in any startup. I learned to defend my specific decisions without stepping on everyone else’s.
• Maze as a quick testing tool is brutal: in 2 days you have 30 responses with click maps and time on screen.

What I would change

• Define my role better before starting. We spent too long deciding “who does what” at every phase.
• Test with real students in schools, not just university classmates. The sample was biased toward people who already know what UX is.
• Narrow the product’s scope. We tried to cover organization plus emotional support plus AI tutoring. Any one of the three on its own would have been deeper.