Prompt Engineer & Conversational Designer for AI Dialogues

Jobtailor; Austin (TX); On-site; USD 100,000 - 170,000; Full time

Job summary

Jobtailor in Austin, TX is seeking a hands-on Prompt Engineer to design multi-turn dialogue flows for voice and text agents. You will define agent personality, write system prompts, and collaborate with product and engineering teams to embed prompts within technical constraints.

The role emphasizes testing with real data, identifying hallucinations and tone issues, and iterating to improve AI system quality across healthtech contexts and regulated environments.

Qualifications

• Hands-on experience in designing complex AI prompts and systems.
• Experience wiring prompts into product workflows with engineers.
• Familiar with testing prompts across real recordings and simulations.

Responsibilities

• Design multi-turn dialogue flows for voice and text agents.
• Define agent personality and maintain consistency across agent family.
• Write, version, and maintain system prompts, conversational flows, and templates.
• Build harnesses and scripts for prompt testing.
• Test prompts using real recordings, transcripts, and simulations.
• Identify hallucinations, incorrect tone, dead ends, and token waste.
• Iterate on prompts and close quality gaps.
• Partner with product managers to translate requirements into agent behavior.
• Work with engineers to integrate prompts within technical constraints.
• Support compliance and privacy obligations.
• Share findings to improve practices across teams.

Skills

Prompt Engineering

AI System Design

LLM API Familiarity

Python Programming

Cross-Functional Communication

Tools

OpenAI

Anthropic

LangChain

VAPI

Job description

Jobtailor in Austin, TX is seeking a hands-on Prompt Engineer to design multi-turn dialogue flows for voice and text agents. You will define agent personality, write system prompts, and collaborate with product and engineering teams to embed prompts within technical constraints.

The role emphasizes testing with real data, identifying hallucinations and tone issues, and iterating to improve AI system quality across healthtech contexts and regulated environments.