These are demos and tools I built to show, concretely, what an accessible and trustworthy conversational product looks like. They all use simulated data and can be tried without signing up. At the end you'll find the link to my portfolio of client work.
Self-assessment · Web
Digital trust check
A free self-assessment of security, accessibility and responsible AI in 15 questions. It returns an instant score and prioritized gaps, no sign-up. It's the check that tends to stall deals before the client's review even arrives.
Open the check →
Conversational prototype · WhatsApp
Anti-fraud prototype for older adults
An automated assistant with human oversight that helps older adults verify whether a message from "the bank" is legitimate. A demo with simulated data, large type and a conversation that meets WCAG 2.2 AA. It shows how trust gets designed for the people who need it most.
See the prototype →
Conversational prototype · WhatsApp
Vera: building-association assistant
A WhatsApp assistant for self-managed buildings: collecting dues, keeping maintenance requests in order and clear communication, without hiring anyone. An interactive demo with fictional data where you pick a scenario and walk the flow, plus a board panel that separates what Vera resolves on her own from what needs your attention.
See the prototype →
Training simulator
Anti-scam simulator: the impostor scam
A practice mode to learn to spot the classic impostor scam with no risk. You receive messages just like a real scammer's and, after each one, an anti-scam guide shows you the red flag and what to do. You choose how to respond and learn by practicing.
Practice →
Training simulator
Anti-scam simulator: prize / lottery
The same practice, applied to the "prize or lottery" scam: that message announcing you've won something that turns out to be a trap. Six signals to detect it, with feedback after each message. No sign-up and no risk at all.
Practice →