Agent Experience Designer, Agentic Voice

Dialpad•San Ramon, CA

22h•Onsite

About The Position

Dialpad is leading the shift to Agentic AI, where intelligent agents not only analyze conversations but also take action by automating workflows, resolving customer issues, and accelerating revenue in real time. The DAART initiative (Dialpad Agentic AI in Real Time) is redefining what a communications platform can do. At Dialpad, AI is integral to how employees do their best work, empowering them to move faster, think bigger, and achieve more. The company values employees who are intensely curious and hold themselves to a high bar, embodying traits like Scrappy, Curious, Optimistic, Persistent, and Empathetic.

Requirements

5+ years of dedicated experience shaping voice user interfaces (VUI), character writing, conversation design, or complex conversational/agentic systems.
Bachelor's degree in Linguistics, Communication, Psychology, Design, or equivalent practical experience.
Demonstrated experience shaping voice user interfaces (VUI), character writing, or complex conversational/agentic systems.
Fluency with LLM-based agent behaviors, prompt engineering, and prompt orchestration (knowing how design choices alter model outputs without relying on code).
Fluency with Text-to-Speech (TTS) controls, including voice selection, SSML tuning, pacing, and emphasis to set broad platform standards.
An exceptional portfolio that highlights voice systems, written persona standards, and interactive logic rather than just static flow diagrams.

Nice To Haves

Experience in regulated, high-stakes verticals (e.g., healthcare, financial services, legal) is a strong plus.
Strong taste and an ear for dialogue—the ability to articulate a character on a page and translate it into consistent AI behavior under pressure.

Responsibilities

Own the agent's global voice, character, and personality, maintaining personal consistency across every vertical.
Own the standard handoff patterns and design systems, ensuring seamless transitions where context is fully preserved when an agent passes a caller to a human.
Own the universal platform quality bar, defining and measuring Consistency, Fluency, and Latency (CFL) and tying personal decisions directly to core metrics like resolution, containment, and sentiment.
Make the voice palette and establish house standards for pacing, prosody, and emphasis that forward-deployed teams will use to build brand-specific experiences.
Partner with AI engineers to orchestrate behavior, escalation instincts, confirmation patterns, and graceful recovery workflows using advanced prompting rather than rigid dialogue trees.
Research and design for distinct behavioral and emotional user states, ensuring the agent adapts seamlessly whether interacting with a patient disputing a bill or a dispatcher tracing a late delivery.