The RoleThis is a foundational engineering hire. You'll report directly to our CTO and co-founder, and work shoulder-to-shoulder with him on the core product. Umi, our AI researcher, runs natural voice interviews with frontline employees and customers at scale, then turns thousands of conversations into intelligence leaders act on.
It's voice, LLM applications, and data pipelines in one product.
This role covers the best parts of the modern AI stack, and your fingerprints will be on all of it.
What You'll Do- Build real-time voice agents (LiveKit, Pipecat, Twilio, Deepgram, ElevenLabs) that hold natural interviews at scale
- Design agentic LLM workflows and agent harnesses on OpenAI, Anthropic, and Google Gemini models: prompts, structured outputs, and evals
- Ship full-stack features end-to-end in Python/FastAPI and TypeScript/React
- Build pipelines that turn thousands of conversations into themes, quotes, and recommendations
- Make non-deterministic systems behave in production: observability, evals, guardrails
- Help shape engineering culture, tooling, and architecture as one of the first engineers
What We're Looking For- Strong engineering fundamentals and experience shipping production systems end-to-end
- Hands-on LLM experience (prompting, structured output, retrieval, evals) beyond demos
- Proficiency in Python and TypeScript, comfort across backend and frontend
- Bias to ship: you own problems end-to-end and iterate fast with real users
- Care for the humans behind every signal
Nice to Have- Real-time voice or audio experience (LiveKit, Pipecat, WebRTC, telephony, STT/TTS)
- Data pipeline or analytics engineering experience
- Early-stage startup experience or 0-to-1 product ownership
Benefits- Meaningful equity and competitive pay
- Medical, dental, and vision for you and your dependents
- Flexible time off and supportive parental leave
- Bright NYC office, team meals, offsites, and customer travel
- Learning budget and top-tier gear
