Manager, AI Engineering – Voice AI
Yum! Center for Global Franchise Excellence
Python
Job Level
Mid-LevelSenior
Tech Stack
Python
About the role
- Design multi turn dialogue flows for text and voice agents with context awareness and memory handling.
- Define personalities and tone across products and agents.
- Create multilingual prompt strategies and develop reusable templates and frameworks to speed development.
- Test prompts with recordings, transcripts, and simulations before and after launch.
- Analyze live interactions to spot hallucinations, tone issues, dead ends, and token inefficiency.
- Iterate quickly to improve quality, safety, and naturalness and use analytics and qualitative insights to measure success.
- Partner with product managers on requirements and goals and work with engineers to integrate prompts within technical constraints.
- Coordinate updates across multiple agents with version control for a unified experience and support compliance and privacy standards.
- Maintain a prompt library with flows, system prompts, and version history and set guidelines for tone, multilingual consistency, and testing.
- Share insights and learnings with the broader product and AI teams.
Requirements
- Proven experience designing multi-turn conversational systems, such as chatbots, voice assistants, or customer service agents.
- Strong skills in prompt engineering techniques (e.g., few-shot, chain-of-thought, RAG pipelines).
- Ability to define consistent, persona-driven conversational styles across multiple agents and products.
- Familiarity with LLM APIs (OpenAI, Anthropic, etc.) and conversational frameworks (e.g., LangChain).
- Excellent writing and linguistic skills, with attention to tone, clarity, and cultural sensitivity.
- Basic programming skills (Python preferred) for testing and prototyping prompts.
- Nice-to-Have: Background in healthcare, caregiving, or other regulated industries.
- Nice-to-Have: Multi-lingual proficiency.
- Role is based in Austin, TX with a hybrid work setup.
