Staff Product Designer, Conversational AI
Deepgram · Remote · San Francisco, CA ·
COMPENSATION
$180k – $240k
Listed on the employer's job board (Ashby)
- Staff level IC role
- Fully remote (PT zone)
- Conversational AI focus
- Define new discipline
- Collaborate with ML/Eng
You will own the end-to-end conversation design for Deepgram's voice agents, defining persona, turn-taking, and recovery strategies. Working directly with machine learning and engineering partners, you will translate behavioral tradeoffs into production-grade features. The role requires establishing evaluation frameworks and developer guidance to ensure natural, high-quality interactions at scale.
What you'll do
- Define persona and voice systems for consistent agent experiences
- Design turn-taking, barge-in, and end-of-turn behaviors with engineering
- Create conversational repair and confirmation strategies for high-stakes actions
- Manage latency-aware pacing including brevity and backchanneling
- Establish quality evaluation metrics and transcript review practices
- Build reference agent experiences and developer-facing design guidance
- Partner with research teams on ASR and TTS behavior criteria
- Set conversation design principles and shared vocabulary for the team
What you'll need
- Deep experience designing conversational behavior for LLM-based voice agents
- Fluency with speech pipelines from ASR through LLM to TTS
- Track record of shipping turn-taking and interruption patterns in production
- Experience building quality measurement loops like evals or benchmarks
- Exceptional writing craft for dialogs, guidance, and documentation
- Ability to influence engineering and ML partners without direct authority
- Experience designing for developers via APIs, SDKs, or documentation
- Working AI practice with clear views on tool capabilities and limits
Nice to have
- Experience with named assistant or production voice agent platforms
- Practice with Wizard of Oz testing and sample dialog methods
- Hands-on work with eval tooling for LLM or voice quality
- Background in high-stakes or regulated conversation domains
- Multilingual or cross-locale conversation design experience
- Experience in high-growth B2B companies with self-serve motions
Stack
- Voice Agent API
- ASR
- TTS
- LLM
- Evaluation tooling
How you'll work
- Fully remote role
- Must align with Pacific Time zone
- Preference for San Francisco location
- Fast-paced AI-first operating rhythm
- Expectation to actively use advanced AI tools daily