← back
Make your LLM app a Domain Expert: How to Build an Expert System — Christopher Lovejoy, Anterior
Takeaway
In vertical AI, the surrounding system encoding domain expertise — not the model — determines who wins the last mile from 95% to 99%.
Summary
- Anterior automates US health-insurance prior authorization (Florence agent, covers 50M lives); domain-native system beat raw model gains, taking accuracy from ~95% to ~99%.
- 'Last-mile problem': specialized industries (e.g. determining 'unsuccessful conservative therapy for six weeks') hide ambiguity that requires a domain expert PM at the system's center.
- Build a customer-specific Adaptive Domain Intelligence Engine: measure with one or two business metrics (e.g. false approvals) plus a failure-mode ontology (record extraction, clinical reasoning, rules interpretation).
- Domain experts must label traces alongside engineers — isolated AI-trace review misses key failure modes.
llm-appshealthcareevals
Original description
Vertical AI is a multi-trillion-dollar opportunity. But you can't build a domain-expert application simply by grabbing the latest LLMs off-the-shelf: you need a system for codifying latent insights from domain experts and using that to drive development of your application. In this talk, we'll describe the system we've built at Anterior which has enabled us to achieve SOTA clinical reasoning and serve health insurance providers covering 50 million American lives. We'll share: - how and why to encode domain-specific failure modes as an ontology - a practical system for converting domain expertise into quantifiable eval metrics - how we structure work and collaboration between our clinicians, engineer and PMs - our eval-driven AI iteration process and how this can be adapted to any industry ---related links--- https://x.com/chrislovejoy_ https://www.linkedin.com/in/dr-christopher-lovejoy/ https://chrislovejoy.me/ https://www.anterior.com/