The Role

We built a voice AI agent that talks to patients about their medical bills and it works. Kora is live with real clients and growing. Now we're expanding into chat, SMS, and outbound calls with a team of five engineers. This role is full-stack in the truest sense: you'll build data integrations, engineer LLM prompts, configure infrastructure, and ship features that real patients interact with the same week.

As a member of the Kora squad (formerly known as Support), your day might look like:

• Driving Innovation: Experimenting with and launching GenAI-powered features in collaboration with AI scientists.
• Harnessing Generative AI: Engineering development and production workflows by integrating LLMs, providing necessary context and constraints using modern practices, and evaluating output for further refinement. This includes context building, prompt design, and building evaluation pipelines that feed directly into Kora's live production system.
• Advancing Patient Understanding: Enhancing Kora, our first-in-class voice AI agent, to deliver personalized billing guidance via empathetic and compassionate communication. This means building the infrastructure that exposes insurance information, billing data, and third-party integrations to Kora and prompting the agent to utilize that data accurately in real-time patient conversations.
• Working Close to Our Users: Actively reviewing call transcripts and conversation logs to identify good and bad product behavior, then feeding those insights directly back into the system. This team runs unusually close to its end users that is part of how we move fast.
• Expanding Kora's Reach: Contributing to the transition from a primarily inbound voice-based agent to a multi-modal platform, including outbound calls, chat, and SMS. There are firsts still to be done on this roadmap.
• Inspiring and Leading Others: Actively contributing to team chartering, managing technical projects, and mentoring peer engineers to take ownership of their own projects.
• Raising the Technical Bar: Taking creative initiative, such as evaluating new approaches to agent-based engineering or developing rapid prototyping techniques. This includes working with the declarative AI agent platform the team uses to build Kora, and adapting quickly to a non-traditional codebase where prompting an LLM replaces hard-coded logic.

What We Look For

• Strong technical leadership with a track record of driving measurable impact across the stack (4–6 years of experience preferred, but not required). 
• Experience in Generative AI and Large Language Models, either via use to increase productivity and efficiency, or by integrating APIs into a product. Personal or professional — we count both.
• Comfort crossing typical engineering boundaries. This role spans data pipeline and ETL work, third-party systems integration, cloud infrastructure setup (AWS or GCP), and product-facing feature development. We are not looking for a specialist, we are looking for someone who adapts to wherever the problem is.
• Demonstrated proficiency in independent, hands-on coding and problem-solving, including handling complex edge cases, optimising for efficiency, and delivering robust solutions within time constraints.
• Excellent communication skills, both written and verbal, with the ability to articulate complex technical ideas clearly, concisely, and with compelling examples, as well as asking insightful clarifying questions.
• A strong commitment to engineering best practices, including writing clean, maintainable, and well-tested code, and a proactive approach to managing technical debt.
• Entrepreneurial mindset; you thrive in fast-paced, ambiguous environments and love solving complex problems.
• Experience with modern languages and frameworks; Python and TypeScript are the primary languages on this team (approximately 50/50). React experience is useful. We are polyglot-friendly, if you learn languages quickly, the specific stack matters less than the depth of what you have built.
• Comfort working with non-deterministic systems. LLM outputs vary; the same code can return different answers on different runs. This is a feature of the domain, not a bug. You will need to design for it, monitor for it, and iterate against it without the safety net of purely deterministic testing.
• Passion for our mission.  You care about making healthcare simpler, fairer, and more transparent for patients.

Compensation Range and Benefits:

• Salary Range*: $170,000-215,000
• This role is equity eligible
• This role offers a competitive benefits and wellness package

*Subject to location, experience, and education