I build AI products and ship them to paying customers. Founding PM at Voltade in Singapore, engineer for five years before that. The work: find the job, build the agent, put it on the customer's WhatsApp, and own what happens next.
Conversational CRM for SMEs. WhatsApp-first inbox where a per-tenant agent triages, drafts, and replies; humans approve. 100+ active SME deployments, 230K+ AI interactions/day.
Self-serve, multi-tenant agent platform, in build now. Each tenant gets an agent-native WhatsApp helpdesk: inbox, event-driven agent harness, self-improving knowledge drive, automation. One pooled deploy; creating a tenant is an INSERT, not an orchestration.
White-label, multi-tenant gym operating system: member PWA, staff console, and platform admin in one app, with Postgres row-level security doing tenant isolation and live MindBody, Telegram, email, and push integrations. Migrated 1,284 workouts and a 441-member CrossFit gym off incumbent software; first commit to gym-wide production in seven weeks.
A couples app for people who hate couples apps. React Native + Expo + Supabase, with a daily prompt drop, a server-gated reveal enforced in row-level security rather than the client, and AI conflict mediation. Live on the App Store since July 2026; two review rejections and the fixes are logged in the repo.
Three posts that capture how I think about AI products: where model safety ends and deployment safety begins, the evaluation framework underneath it, and the system behind the headline numbers.