Build a Stateful AI Agent for Event Sponsorships

Description and use

Build a stateful AI agent for event booths that maintains per-attendee conversation state, qualifies leads, instantly alerts sales, and schedules follow-ups, delivering continuous data flow and improved sales effectiveness.

Problem

Event booths generate many conversations, but useful context is lost when attendees leave, leading to fragmented data and missed sales opportunities.

Outcome

The agent consolidates fragmented event interactions into a durable, cross-channel workflow, enabling real-time lead routing, automated follow-up, and concise attribution reporting, thereby improving sales engagement and data quality.

Workflow

• Serve branded microsite and chat API
• Detect attendee language and store ISO code
• Guide attendee through qualification flow (name, company, use case, size, timeline, preferred channel)
• Upsert each response into SQLDB and actor state
• Recognize intents (product question, demo request, giveaway)
• For product questions, call Telnyx AI Inference to generate answers
• When demo requested, send immediate SMS alert to sales with captured context
• Schedule follow-up via Agent SDK based on preferred channel
• Send or log follow-up message when scheduled task runs
• Provide GET /api/report endpoint for lead attribution
• Deploy locally with in-memory KV, SQLite, and stubs, then ship to Edge
• Configure secrets and enable real communications after testing

Evidence

The event-sponsorship-agent example turns that fragmented workflow into one stateful application on Telnyx Edge Compute.

Source

When an attendee completes the qualification fields and requests a demo, the agent sends an SMS alert to the sales team immediately.

Source