BIKIN INGAT
An AI-powered medication adherence app built for the aging population. Built with Team 404 — Tatiana Fyka Binti Azman, CK Yong, Amelia Michelle Bernard, Calista Tatiana Lo, and Aquino Pentojo Joikon — during Gamuda AI Academy Sabah, Cohort 3.
How it works
The app removes manual data entry from medication tracking. A user scans a prescription or a medicine box; Google Cloud Vision reads the label, and a Gemini-powered RAG layer cross-references the extracted drug information against medical reference data before returning it in plain language. Firebase persists the resulting schedule and user data, so reminders and dosage history survive across sessions rather than resetting each time the app is reopened.
Core features
Scanning a prescription or medicine box auto-builds a dosing schedule — no manual data entry required. Medications are explained in plain language rather than clinical shorthand, and the system flags drug interactions before they become a problem instead of leaving that check to the user. When a refill is needed, the app connects the user to the nearest pharmacy, closing the loop between "notice you're low" and "actually get more."
Why it works
Medication non-adherence in older adults usually traces back to two failure points: prescriptions that are hard to read or understand, and no reliable reminder system once the bottle's put away. Scanning instead of typing removes the first point of friction; plain-language explanations and automatic interaction checks remove the second layer of risk — misunderstanding what you're taking. And because it's tied to a real refill path, the system doesn't just track adherence, it helps sustain it.
Key Features & Details
- —Scans prescriptions and medicine boxes to auto-schedule doses
- —Explains medications in plain language and flags drug interactions
- —Connects users to the nearest pharmacy for refills
- —Built on Google Cloud Vision, Gemini, RAG, and Firebase

