About the role
Our client is an AI health-tech company building voice agents that take on healthcare’s staffing shortages and admin burden. Their agents talk with patients every day, booking appointments, answering billing questions, handling refills and checking in after hospital visits, for health systems, provider groups and FQHCs.
You’ll own AI agents end to end, from conversation design to production, and use live patient conversations to make them better.
- Design, build and ship AI agents end to end: conversation flows, prompts, tool calling, retrieval and the backend services behind them
- Build the eval suite (offline tests, simulated conversations, production monitoring) that every agent change has to pass before it ships
- Improve instruction following, tool-calling reliability, latency and cost across the LLM and speech stack
- Help shape the next-generation voice agent platform
- Work with product, clinical and customer teams to turn workflows like scheduling, refills and billing into reliable agent behaviour
What’s in it for you:
- Real ownership: ship multiple production agents with health system partners in your first three months
- A mission that matters: making healthcare work better for patients
- Direct access to the SVP of Engineering on a small team where your work reaches production fast
$160-210k | Hybrid, NYC or SF
Who you are:
- 5+ years in software engineering, including 1+ year shipping LLM-powered apps in production
- Strong in Python and/or TypeScript, comfortable across backend APIs and data
- Hands-on with prompting, function calling, RAG, structured outputs and agent orchestration
- Experience designing evals and using data to decide what to fix next
- Product sense: you care whether the patient got their appointment, not just whether the metric moved
- Voice AI (STT, TTS, turn-taking, Twilio/SIP/WebRTC) or healthcare/HIPAA/FHIR experience is a plus
If you want to point at an agent a year from now and say “a patient talked to that every day, and I built it”, let’s talk.
Apply below or reach out directly at Julia at Alldus dot com.
Requirements
- 5+ years in software engineering, including 1+ year shipping LLM-powered apps in production
- Strong in Python and/or TypeScript, comfortable across backend APIs and data