Personal product / AI learning for children
EpisodeIQ
Kids ask any question and get an illustrated, read-aloud explanation matched to their level, followed by a short quiz.
- Status
- Built and deployed
- Year
- 2026
The live product, embedded. If it does not load here, use the button above to open it.
My role
- Product concept, interface design and development.
- Designed the parent flow where a parent creates the child's profile.
Children ask questions constantly, and the answers they find online are rarely written for them.
Why it mattered
A clear, friendly explanation with pictures and a voice makes hard ideas approachable, and a quiz shows whether it stuck.
- Answer in the child's level of language, not an adult's.
- Combine pictures, read-aloud audio and a quiz so the same idea is met three ways.
What it does
- Explanations matched to the child's school board and grade, in English, Hindi or Marathi.
- Scene pictures and browser read-aloud.
- Quiz after each explanation.
- Parent-created child profiles.
What I wanted to build: A safe place where a child can ask any question and get a clear, picture-led explanation that matches their school level, then check what they learned.
How it came together
- Wrote the requirement: parent sets up the child, child asks, the app explains, then quizzes.
- Designed the prompts so the explanation follows the child's board and grade, and can be written in English, Hindi or Marathi.
- Built the screens, then iterated on layout and tone with the AI tools until it felt child-friendly.
- Connected authentication, a database and the AI model, then added scene pictures and read-aloud.
- Tested the flows, fixed issues and deployed to Vercel.
What I did
- Defined the requirements and the user journey before any code was generated.
- Made the design and product decisions, and reviewed what the tools produced.
- Tested, debugged and decided what to keep, rewrite or cut.
- Configured the services and handled deployment.
Where AI tools helped
- AI tools drafted and refactored code from structured prompts, and I iterated on the results.
Services connected
- Clerk for parent sign-in and protected dashboard and setup pages.
- Supabase as the database, accessed from server routes.
- Groq language model to generate episodes and quizzes from structured prompts.
- Image URLs generated per scene through an image generation service.
- Browser speech synthesis for read-aloud, with Hindi and Marathi voices where the browser provides them.
Working today
- Parent sign-in, child profiles and a child login.
- Episode generation with scenes, key points and pictures.
- Quiz generation and submission.
- Read-aloud in the browser.
Still prototype or planned
- Payments and email packages are installed in the project but are not wired into any feature.
- Not tested at scale and not presented as production-hardened.
Integrations above were checked against the source code.
- Keeping explanations accurate, simple and safe for children.
A deployed, working app you can try from the preview above. No user numbers are claimed.
Child-friendly design is mostly about removing things.
Want a friendly, focused learning product?