
2
1
Indonesia
5+ years of experience
AI Agent Tooling · Full-Stack Developer · Educator. Building practical AI systems, developer tools, and education technology. Contribute to open-source AI agent infrastructure (ZizkaDB, openclaw). Build full-stack apps with Next.js, TypeScript, Supabase, PostgreSQL. Educator at MAM 1 Paciran Lamongan — teaching digital literacy and software engineering. Focused on AI tools for Indonesian education and student communities.
.png&w=828&q=75)
MeetingIQ is an AI-powered meeting preparation assistant that helps sales and account managers walk into every customer meeting fully prepared. THE PROBLEM: Sales teams waste 30+ minutes before each customer meeting researching the company, digging through scattered notes, and trying to remember past interactions. Context is lost between meetings, and new team members start from zero. THE SOLUTION: MeetingIQ combines three powerful AI capabilities into one seamless workflow: 1. LIVE WEB RESEARCH — Uses Exa neural search API to gather real-time company intelligence, news, and market data. 2. PERSISTENT MEMORY — Leverages Mem0 long-term memory layer to remember past interactions, preferences, and context for each customer across meetings. 3. 30-SECOND BRIEFS — Synthesizes research and memory using Google Gemini 2.0 Flash to generate concise, actionable meeting briefs. KEY FEATURES: - Real-time company research via Exa neural search - Per-customer memory that survives across meetings (Mem0) - AI-generated meeting briefs with talking points and conversation starters - Meeting outcome logging that feeds back into memory - Company profiles with contacts and activity timeline - Supabase PostgreSQL with Row-Level Security TECH STACK: - Next.js 16 with App Router and TypeScript - Tailwind CSS 4 with dark theme - Supabase (PostgreSQL + RLS) - Exa Neural Search API - Mem0 Long-term Memory API - Google Gemini 2.0 Flash - Deployed on Vercel Built for the AMD Developer Hackathon: ACT II — Unicorn Track.
13 Jul 2026