About the role
Insurance runs on documents: ACORD forms, schedules of values, loss runs, policies, endorsements and email threads. As an Applied AI Engineer on Document Intelligence, you will build the models and pipelines that read them accurately and explain where every extracted value came from.
What you'll do
- Build extraction and classification pipelines for submissions, policies and loss runs
- Combine LLMs, layout models and rules to hit production accuracy targets
- Create labelled datasets and evaluation suites with our insurance specialists
- Ship citation and confidence features so reviewers can trust every field
- Track new research and turn the useful parts into product quickly
What we're looking for
- 3+ years building ML or LLM systems that run in production
- Strong Python and experience with document AI, OCR or information extraction
- A rigorous approach to evaluation and error analysis
- Pragmatism about when a simple rule beats a model
- Experience with insurance or financial documents is a plus
About Clarix
Clarix builds agentic AI for insurance operations. Carriers, MGAs and brokers use Clarix agents to take on submission intake, FNOL, policy checking, certificates, endorsements, loss runs and broker servicing, working inside the core systems their teams already use, with every decision explained. We are a small, senior team backed by leading investors, and we ship to some of the most demanding operations teams in the industry.