Applied AI Engineer

Exactrx is fixing one of the most broken parts of US healthcare: medical billing. Providers spend over $250B a year just to get paid by insurance, and most of that work still runs on people tracking payer rules by hand. Exactrx is replacing that with AI agents that log into the systems providers already use, code and submit claims, and get them paid. As the Applied AI Engineer, you’ll own the intelligence behind every claim the company submits, from the agents that read clinical notes and assign codes to the payer logic that catches problems before they become denials. You’ll be one of the first engineering hires outside the founding team, reporting to the CTO and Head of AI, with real ownership from day one and a direct line to customers seeing the results. It’s a rare chance to ship AI that moves real dollars in a $250B problem, backed by Overline, Tech Square Ventures, and Northwestern Medicine.

ABOUT THE ROLE

As our Applied AI Engineer, you'll own the intelligence behind every claim Exactrx submits: the agents that code encounters, the payer logic that catches problems before submission, and the analysis that shows customers what changed. You'll report to our CTO + Head of AI, and work daily with our service delivery team, whose coders review the cases our AI agents can't yet handle.

You'll own whether our agents work in production, from framing the problem through evaluation, deployment, and the failures that make the next version better. Within 90 days, you'll own autocoding, payer logic, and analytics. You'll be one of our first engineering hires outside our leadership team, so we need a self-starter who takes ownership, drives execution, and works closely with leadership to align on priorities and key decisions.

WHAT YOU'LL DO

• Build and improve LLM-driven agents that read clinical documentation and assign CPT, ICD-10, and modifier codes.
• Turn payer rules and published policies into claims checks that run before submission, and grow our claim rules library as new denials appear.
• Build evaluation loops from coder feedback, remittance data, and production traces to measure accuracy, catch regressions, and choose between prompting, retrieval, rules, and fine-tuning.
• Analyze claims data for new customers to determine drivers of AR days and denials in dollars, and build report cards for our sales team.
• Track performance after go-live.

WHAT WE'RE LOOKING FOR

• 3-5 years in software engineering, ML engineering, or applied AI, including at least 1 year shipping AI systems to production users.
• Strong Python background, and experience building LLM applications, agents, retrieval, or structured extraction.
• Experience designing evaluations and benchmarks for AI systems.
• Comfort with PHI under HIPAA controls.
• The ability to explain an analysis to a non-technical customer and defend the numbers.
• Nice to have: medical coding knowledge or a CPC, revenue cycle or payer experience, HL7, FHIR, or EDI experience, web development experience, and early-stage startup experience. Experience with messy real-world data such as 837 and 835 claims files and clinical notes.

Exactrx is an equal opportunity employer. We don't discriminate based on race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, veteran status, or any other protected characteristic. If you need accommodation during the application process, let us know. Background checks are administered in accordance with applicable law.