Ticket types: Billing · Technical · Refund · Complaint · General
Blue bars = AI value · Amber bars = Human value
drag to pan · scroll to zoom
AI-powered customer service works by constantly deciding, ticket by ticket, whether an automated agent should answer or a human should take over — and that decision is itself learned from experience. This 2D dashboard renders that learning loop as a pannable/zoomable flow diagram: five ticket types stream in continuously, an ε-greedy policy chooses AI or Human for each one, a reward is sampled from a hidden satisfaction model that also charges a tunable escalation cost, and the resulting Q-value estimates grow as live bars beside every ticket-type node. A scrolling strip chart below tracks reward and escalation rate over time so you can watch the policy converge — or destabilise — as you tune exploration, learning rate, escalation cost and reward noise.