AI Engineer

Join us as a Founding AI Engineer, where you'll have the opportunity to shape the core AI architecture and define our machine learning strategy. You'll work closely with the CTO and founding team to design and implement LLM-powered systems for automating medical coding, build agent-based workflows, and integrate AI outputs with user-facing apps. This is a full-time role focused on deploying production-ready AI systems in real-world clinical environments.

• Design and implement LLM-powered systems for automating medical coding and build agent-based workflows tailored to complex clinical operations.
• Collaborate on integrating AI outputs with user-facing apps used by doctors and hospital staff, and work closely with hospital IT to build secure and scalable data ingestion pipelines.
• Own the full lifecycle of AI features: research, prototyping, evaluation, deployment, and monitoring, while partnering with the CTO to define the AI roadmap.

We’re looking for someone who’s passionate about the ongoing AI revolution, action-oriented, autonomous, and ambitious. You are the ideal candidate if you have:

• 5+ years of experience working on applied ML/AI problems in production environments
• Strong experience with LLMs and NLP, including prompt engineering and/or fine-tuning
• Proficiency in Python or Node.js and experience with ML libraries (e.g. Hugging Face, LangChain, PyTorch, etc.)
• Familiarity with backend services (Node.js/TypeScript) and data infrastructure is a strong plus
• A deep sense of ownership and ability to move from prototype to product quickly
• Experience working with sensitive or regulated data (healthcare, finance, etc.) is a bonus

**Technical stack**

• Backend: Typescript with NestJS, Express, Prisma, Postgres
• Frontend: React, TanStack, Tailwind
• Data: Python & Node (for low level proxy servers)
• Tools: Monorepo, Github, Github actions
• Infra: AWS, Azure, Cloudflare, Docker, Terraform, Kubernetes
• CI/CD: Github, Github Actions, Monorepo setup
• Observability: Datadog
• AI/ML: Hugging Face, LangChain

**Engineering Mindset**

• Data security is our foundation, given our work with sensitive health data
• We focus on solving user problems, not shipping features — deep product involvement is essential
• Full type safety from database to UI
• Rapid development with tools like Cursor
• Automated best practices with `eslint`, `Prettier`, `Jest`, and `TypeScript`
• We leverage the latest technologies and libraries but sometimes, old boring tech that does the job is what’s needed
• Infrastructure should empower — not block — product iteration