Day 1 Update – Ideation & Problem Framing SIMPEL – Day 1: Finding the Right Problem
We started with a big question: Why do millions of eligible Indonesian families miss out on social assistance programs they desperately need?
It's not because they don't want help. It's because the system is fragmented, the rules are written in dense bureaucratic language, and there's no single place to check eligibility across programs like PKH (cash transfers), BPNT (food aid), PIP (school assistance), and PBI-JKN (free healthcare).
So we decided to build SIMPEL (Sistem Informasi Mudah Pahami Eligibility Layanan — "Simple Information System for Understanding Benefit Eligibility"). An AI-powered navigator that asks 7 plain-language questions and instantly tells you which programs you may qualify for, why, and exactly where to go.
Today we: -Selected our track: Public Services → Benefits Navigator -Grounded the problem in a real user persona: Ibu Sari, a daily-wage worker and mother of two in rural Bogor -Mapped out the 4 national programs we'll cover (PKH, BPNT, PIP, PBI-JKN) -Designed the 7-question flow (income, family size, school children, toddler/pregnancy, elderly, disability, house condition) -Made key architecture decisions: single-file Streamlit app, Random Forest classifier, synthetic data, no login, no paid APIs -Set up GitHub repo and folder structure
Key insight: A structured questionnaire is better than a chatbot for our users — many have limited literacy, slow 2G connections, and just want clear answers fast.
Tomorrow: building the app skeleton and training the AI model.
Log in or sign up for Devpost to join the conversation.