Prompt/LLM Engineer

Why work at nimble?

This is a great opportunity to join a well-established and market-leading brand serving a high-growth end market while gaining valuable experience working closely with Executive leadership. As an organization, we are in high-growth mode through acquisition with a laser focus on positive culture building!

Who we are!

nimble solutions is a leading provider of revenue cycle management solutions for ambulatory surgery centers (ASCs), surgical clinics, surgical hospitals, and anesthesia groups. Our tech-enabled solutions allow surgical organizations to streamline their revenue cycle processes, reduce administrative burden, and improve financial outcomes. Join over 1,100 surgical organizations that trust nimble solutions and its advisors to bring deep insights and actionable intelligence to maximize their revenue cycle.

What you'll do!

Design and ship prompt-driven and retrieval-grounded language model features that make complex revenue cycle data easy to use. You will work hand in hand with our existing AI team to create reliable, explainable outputs that people can trust in their daily work.

Why this role matters!

Your work will help users summarize documentation, understand denials, find answers in data, and complete tasks faster, while protecting privacy and keeping costs in check.

What you will do:

• Write and refine prompts and simple orchestration logic that deliver consistent, verifiable results
• Build grounded flows that read from approved data sources and return answers with citations or evidence
• Create test sets and automatic checks for accuracy, coverage, safety, latency, and cost
• Add basic guardrails to reduce hallucinations and keep outputs within expected formats
• Partner with engineers to integrate LLM outputs into review and approval steps, and with the product to define clear user expectations
• Track usage, quality, and spend, learn from feedback, and improve fast
• Document patterns and help train peers so more teams can adopt them
• Preparing data for use in updating or training machine learning processes
• Collaborate with developers to test and deploy improvements
• Providing and explaining reports regarding AI outcomes and performance

What you will bring

• 2 plus years working with LLMs or conversational systems in production
• Strong prompt design skills and working knowledge of Python for experiments and simple services
• Understanding of retrieval, embeddings, and vector search fundamentals
• Ability to write clear specs, think through failure modes, and balance quality, speed, and cost
• Prompts meet regulatory and compliance requirements

Nice to have

• Familiarity with healthcare RCM concepts, EDI basics, and common documentation patterns
• Experience building chat interfaces over governed data models or BI datasets
• Background with human in the loop review and red teaming

What success looks like

• In 90 days, you have at least one LLM feature in production with automated tests and live dashboards for quality and cost
• In 6 months, answer quality and task completion rates improve meaningfully, while unit cost goes down
• Ongoing, you establish reusable patterns and work closely with the AI team to raise the overall bar