Build the infrastructure that connects personal agents to healthcare organizations.

About Doe

Patients are starting to reach healthcare through their own personal AI agents: to find a provider, book a visit, ask about coverage, or check on a result. Doe is the infrastructure that connects those agents to healthcare organizations, from single practices to multi-site groups.

On top of it, we build AI tools for providers in every discipline, from primary care and dentistry to mental health, physiotherapy, and specialty care. Each organization gets its own intelligence model, trained on how it actually runs, and agents that answer, schedule, and follow up on its behalf. We are a small team of engineers, designers, and clinicians building it together.

The Role

Doe sits between patients' personal agents and the healthcare organizations they reach. You will build the systems that carry every request across that line, and the agents that answer, book, and follow up for providers: rapid, reliable, and observable.

You will own the path from a patient's first word to the provider's record, for every discipline we serve, and make it hold up at scale.

What You'll Do

- Build and operate the runtime for voice and text agents:



orchestration, tool calls, memory, and retries.
- Design the interfaces personal agents use to reach healthcare organizations through Doe.
- Bring latency down and reliability up across speech, model, and telephony providers.
- Build evaluation and monitoring so every agent change is measured before and after it ships.
- Design how each organization's intelligence model is trained, versioned, and served.
- Own incidents end to end, and make sure they do not happen twice.

What You'll Bring

- Four or more years building distributed systems or production infrastructure.
- Hands-on experience shipping LLM-based systems to real users.
- Strong fundamentals in Python or TypeScript, queues, and cloud infrastructure.
- The judgment to make systems simple before making them clever.

Nice to Have

- Experience with real-time voice, telephony, or WebRTC.
- Work on evaluation frameworks for language models.
- Experience with healthcare data, such as FHIR or HL7, or another regulated industry.

Doe One

The most competitive benefits package among Toronto startups.

#J-18808-Ljbffr

📌 Founding AI & Infrastructure Engineer (Full-Stack) (Canada)
🏢 Doe Intelligence
📍 Canada