CX Conversational Designer - AI Agent Dialogs (Remote)

Flock; United States; À distance; USD 85 000 - 100 000; Plein temps

Avantages offerts par ce poste

Stock options

Résumé du poste

Flock is seeking a Conversation Designer to own the craft of AI support agent conversations from end-to-end. You’ll design intents, train utterances, and deploy prompts that drive a measurable lift in AI-resolved conversations while ensuring safe handoffs for trust-critical escalations.

You’ll define guardrails for SEV incidents and VIP/Safe City accounts, build a analytics-backed backlog, and codify standards so the agent portfolio scales without sacrificing quality.

Qualifications

• Experience designing and iterating conversational flows in live AI agent platforms.
• Ability to map intents, train utterances, and implement fallback strategies.
• Hands-on prompt development for LLM-based agents, not legacy bots.
• Fluency in reading logs and performance dashboards to diagnose failures.
• Clear, concise writing for customer-facing bot voice.

Responsabilités

• Own end-to-end conversation design for Flock’s AI support agents from intent mapping to production deployment.
• Achieve measurable lift in AI-resolved conversations with high-quality handoffs for trust-critical escalations.
• Design escalation and fallback logic for SEV incidents, VIP, and Safe City accounts to a human review.
• Build a conversation analytics practice to turn data into a prioritized design backlog.
• Create documentation and standards to scale the agent portfolio without quality loss.
• Partner with Support, Product, and Trust to align behavior with SLA and safety requirements.

Connaissances

Intercom/Fin platform

Conversational design

Prompt development

Reading logs & dashboards

Description du poste

Flock is seeking a Conversation Designer to own the craft of AI support agent conversations from end-to-end. You’ll design intents, train utterances, and deploy prompts that drive a measurable lift in AI-resolved conversations while ensuring safe handoffs for trust-critical escalations.

You’ll define guardrails for SEV incidents and VIP/Safe City accounts, build a analytics-backed backlog, and codify standards so the agent portfolio scales without sacrificing quality.