Parachute Health is transforming post-acute care as the leading digital ordering platform for medical equipment and supplies. The Staff Software Engineer, Applied AI will design and build AI features, contribute to full-stack product development, and enhance evaluation frameworks to improve product capabilities.
Responsibilities:
- Design and build agentic LLM pipelines that power AI features across the product
- Build full-stack product features: Python/FastAPI web applications, TypeScript or React web tooling
- Prompt-engineer and iterate on LLM prompts and models. Evaluate and recommend improvements based on accuracy, latency, and cost
- Strengthen our evaluation framework: curated datasets, evals, experiments, and regression identification, so we can ship changes with confidence
- Improve production observability: accuracy signals and feedback loops
- Contribute to architecture and technical direction
- Actively partner with other product teams to understand their AI use cases and identify how to best enable and support them
- Advise on when to apply an AI/LLM solution, and when not to: championing AI where it drives speed and ROI and steering away where it doesn't
Requirements:
- 5+ years of professional software engineering experience
- Experience working in a startup, incubator, or other fast-paced 0-to-1 environment where you own features end-to-end and set your own direction
- Strong full-stack web application experience in Python
- Has shipped reliable agentic LLM workflows to production, with any orchestration framework (LangGraph, LangChain, etc)
- Experience evaluating and testing LLM workflows: eval harnesses, golden datasets, regressions, and observability
- Familiarity with prompt engineering and the prompt iteration lifecycle
- Fluent with AI-powered coding tools (e.g. Cursor, Claude Code) to move quickly
- Strong written and verbal communication; able to drive alignment without being asked
- Must reside in the U.S
- Healthcare workflows or health-tech experience
- Ruby on Rails experience
- A track record of evaluating emerging AI tools and driving their adoption on a team