Software Engineer
Summary
Full-stack engineer at Lyrebird Health building its AI scribe for GPs: taking features from idea to production across a SvelteKit front end, Python FastAPI services and Postgres on AWS, with daily hands-on LLM work on prompts, structured output, streaming, and generation latency/cost in a live clinical product.
The role asks for a full stack engineer who can take features from idea to production across a SvelteKit front end, Python FastAPI services and Postgres. A strong candidate brings production TypeScript and Python, hands on work integrating large language models, and solid SQL and AWS operating experience. Breadth across layers and the judgement to lead in some areas and learn in others matter more than depth in one layer.
How it fits the AI era
Much of the work sits directly on AI: building clinical document types, tuning prompts and models, and cutting the latency and cost of generation without losing quality. The human side stays in tracing slow requests, finding root causes of incidents, reviewing code and deciding what is safe to ship. AI speeds up parts of the build, but judgement about quality and safety remains with the engineer.
How it builds AI experience
The role offers hands on practice with prompt engineering, structured output, streaming and cost control inside a live clinical product. It also builds skill in evaluating prompts and protecting personal data in text. Someone preparing could build small projects that call language models with structured output and measure latency and cost, since that mirrors the daily work here.
Worth showing
- TypeScript and SvelteKit
- Python FastAPI services
- LLM integration and prompts
- AWS and observability
This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.
About Lyrebird Health
An AI scribe built directly into Australia's most-used GP software, auto-filling care plans from the consultation rather than just writing a note.
