PPL Spark builds AI tools that help people get better at the conversations they're already having.
Unlike coaching chatbots that take you into a scripted "nowhere-land," our tools work on real material — live calls, recorded meetings, actual team dynamics — and help participants see and improve what's happening in the moment or just after.
Two product lines
1. Live Companion (AI Call Assistant) A real-time meeting companion. It transcribes, identifies speakers, tracks the agenda, notices when someone has been silent, and suggests moves to the host: "Reiner has been quiet for 5 minutes — bring him in." It's a facilitator's second brain.
2. Agent-Based Modeling Feed in a situation (a regulation, a lobbying campaign, a family decision) and the system extracts relevant actors, builds their personas from public information, simulates how they'd react, and produces a report. Used for policy analysis (e.g. MiCA regulation with 49 actors) and personal scenario planning.
Where it's proven
Zurich Kundencenter Lernpilot — a pilot using AI to mirror and improve customer-service conversations in an insurance call center, measured against existing NPS. This is the "shape" of our best-fit customer: defined conversations, measurable quality, real business stakes.
The underlying idea
Conversations in defined systems — organizations, call centers, team meetings — have predictable structure. That makes them tractable for AI to observe and improve. Our edge is the art-of-dialogue lens on top of the tech: we don't just summarize, we help conversations become the kind that produce decisions, insight, and collaboration.