Position Overview
We are seeking an exceptionally talented and highly operational AI engineering Lead to spearhead our Intelligence, RAG & Context stream. You will lead hands-on, by example, write core Agentic AI code, architect production-grade pipelines, and mentor a brilliant team of AI and Data Engineers to deliver high accuracy and explainable AI based solutions.
The ideal candidate has lived through the technological paradigm shift from ML to LLM. You have a deep, foundational background in traditional Machine Learning (training, fine-tuning, statistical assessment) but successfully pivoted your career into LLMs, Advanced RAG, and Agentic AI systems, for the healthcare insurance market. Your mission will be to turn raw cognitive power into concrete, enterprise-grade SaaS features. You will lead the development of AI capabilities that power the claims management process, from data entry and enrichment to FWA (Fraud, Waste, and Abuse) and adjudication, providing claims handlers with relevant, accurate insights to accelerate decision-making.
Key Missions & Responsibilities
Operational & Architectural Leadership: Architect, code, and deploy multi-agent systems and deep learning models for Qantev's Claims Data Platform. You own the technical delivery of your stream.
Bridge the Paradigm Shift: Apply your dual expertise to combine traditional ML (for deterministic tasks like OCR/ICR parsing) and rule based (system expert) with GenAI/Agentic architectures (LangGraph, SmolAgents, MCP) for complex hybrid reasoning and fraud detection.
R&D Industrialization: Ensure that all AI models are wrapped in strict, high-performance, business oriented, API-first contracts that allow seamless integration with our product dev team.
Team Mentorship: Foster a high-ownership engineering culture. Mentor and develop Data Scientists, AI Engineers, and Back-End Data Engineers through code reviews, pairing, and architectural guidance.
Evaluation & LLMOps: Establish rigorous evaluation frameworks. Build automated benchmarking pipelines to track model drift, cost-efficiency, latency, and token optimization before any production release.
