Inspiration
Mental health is one of the biggest silent crises affecting students today — yet it remains one of the most underfunded and overlooked problems in education. 1 in 5 teenagers experiences a mental health condition, but most never seek help. Not because they don't want to — but because the barriers are too high. Therapy is expensive. School counselors are overwhelmed. And the mental health apps that exist today are either too clinical, too expensive, or simply not designed for how teenagers actually live.
I wanted to change that. But what pushed me to actually build it was my own reality — I'm a Grade 12 student in Nigeria with no laptop, no fancy setup, and no team. Just an old Android phone and a belief that where you start should never limit what you can build.
MindCheck was born from one simple question: what if checking in on your mental health was as easy and normal as checking the weather?
What it does
MindCheck is a free daily AI-powered mental wellness companion built specifically for students aged 13–20.
Every day, a student opens MindCheck and:
- Selects their current mood from Awful to Great
- Writes freely about what's on their mind in a completely judgment-free space
- Logs their sleep hours and stress level
- Tracks whether they exercised
The app sends all of this to an AI that generates:
- A personalized wellness score from 1 to 100
- A warm human assessment of how they're doing
- 3 specific actionable tips tailored to their exact situation that day
- A small daily challenge to build positive habits
If the AI detects a student may be in crisis, MindCheck automatically displays the 988 Suicide and Crisis Lifeline because student safety is never optional.
Students also track their progress over time through a weekly bar chart, full check-in history, and a streak counter that rewards daily consistency.
MindCheck is already live and free at mindcheck.tiiny.host
How i built it
MindCheck is built entirely with vanilla HTML, CSS, and JavaScript — every single line coded on my Android phone using Acode, a mobile code editor. No laptop. No desktop. No development environment. Just a phone and determination.
The AI layer is powered by OpenRouter's API connected to the arcee-ai/trinity-large-preview model. I engineered a structured prompt that forces the model to return clean JSON — with the wellness score, assessment, tips, daily challenge, and crisis flag all parsed dynamically into the UI.
All user data is stored locally using localStorage, meaning zero data collection and complete privacy for every student who uses it.
The design was built mobile-first with a deep dark theme, animated background orbs, a live ticking clock, and smooth transitions — because a mental health app should feel calm, safe, and beautiful to open every single day.
Challenges i ran into
Coding on a phone is genuinely hard in ways most developers never experience. No split-screen debugging. No browser dev tools. No reliable autocomplete. Every bug had to be spotted manually by reading through code on a small screen and fixing one mistake could take 3x longer than on a laptop.
Multiple AI models I tried including LLaMA, Mistral, and Gemma returned endpoint errors mid-development, forcing me to research, test, and switch models multiple times under deadline pressure.
Getting the AI to consistently return structured JSON required deep prompt engineering — I had to be extremely precise about field names, data types, and response format to make parsing reliable.
Every single challenge was solved on a 6-inch screen. 📱
Accomplishments that i'am proud of
- Built and deployed a fully functional AI wellness app entirely on an Android phone — no laptop, no desktop
- Engineered a structured AI prompt that reliably returns personalized JSON parsed into a rich dynamic UI
- Designed a mobile-first interface that feels premium, calm, and safe — something a struggling student would actually want to open every day
- Built automatic crisis detection that surfaces the 988 Lifeline for users who may be in danger
- Zero data collection — all data stays on the user's device, fully respecting student privacy
- Shipped a live deployed product in under 2 weeks, completely solo, with no laptop
What i learned
I learned that constraints are not excuses — they are teachers. Not having a laptop forced me to plan every feature carefully before writing a single line of code. I could not be sloppy because fixing mistakes on a phone costs twice the time and energy.
I learned how to work with AI APIs under real pressure — handling model availability failures, engineering reliable structured prompts, and building graceful fallbacks when responses do not parse as expected.
Most importantly I learned that the best products are built with empathy first and technology second. Every decision in MindCheck — the warm AI tone, the calming dark theme, the crisis detection, the judgment-free text box — was made by asking one question:
"How would a struggling 15-year-old feel using this?"
What's next for MindCheck
MindCheck is not just a hackathon project. It is the beginning of something much bigger.
Phase 1 — Scale the user base Keep MindCheck completely free for all students worldwide. Focus on impact first, sustainability second.
Phase 2 — School partnerships Partner with schools and districts who pay a subscription for anonymous aggregate wellness data — helping counselors identify at-risk periods like exam season without ever compromising individual student privacy.
Phase 3 — Premium features Optional paid features for students who want more: guided breathing exercises, AI coaching conversations, and personalized mental health plans. Basic access will always remain free.
Phase 4 — Expand the platform
- Mood pattern detection that proactively reaches out when a student has been consistently struggling
- Parent and guardian dashboard with optional consent-based sharing
- PWA support with daily check-in push notifications
- Multilingual support starting with African languages
The global mental health app market is worth over $5 billion and growing fast. But almost nobody is building for teenagers in developing countries. That is MindCheck's market. That is my mission.
I built this on a phone. Imagine what I'll build with real support behind me. 🚀
Built With
- acode-editor(andriod)
- css
- html
- javascript
- localstorage
- openrouter-api
- rcee-ai/trinity-large-preview
- tiiny.host
Log in or sign up for Devpost to join the conversation.