AI Integration Engineer

About the Role

ClaimOS doesn't train models. It orchestrates them and verifies their output. You'll build the layer that makes AI claims decisions trustworthy: multi-model consensus, confidence scoring, hallucination detection, and model-agnostic abstraction so the platform improves as models improve.

What You'll Build

• Multi-model consensus pipelines for parallel adjudication across LLMs with agreement-based validation
• Confidence scoring framework that determines when to trust AI output and when to escalate
• Hallucination detection for claims-specific outputs (coverage determinations, benefit calculations, exclusion matching)
• Model-agnostic abstraction layer to swap underlying models without changing verification logic
• Prompt engineering and evaluation pipelines for insurance-specific adjudication tasks
• Integration with model providers (OpenAI, Anthropic, open-source) with fallback and failover

Requirements

• Technical:

  • Strong Python and/or Java/Kotlin experience
  • Hands-on experience with LLM APIs and prompt engineering
  • Understanding of model evaluation, scoring, and output validation
  • Familiarity with RAG patterns, embeddings, and document processing
  • Experience with ML pipeline orchestration (preferred)

• Experience:

  • 3+ years building systems that use AI/ML models in production
  • Experience with multi-model or ensemble approaches (advantageous)
  • Insurance or regulated-industry background (advantageous)

What We Offer

• Competitive salary with Virtual Stock Option Plan (VSOP)
• Remote-first, async work culture
• Work on a novel verification architecture, not another chatbot wrapper
• Conference and learning budget

How to Apply

Submit your application below with your CV, and any relevant links (GitHub, publications, side projects).

We are an equal opportunity employer.