Underwriting AI & Data Analyst

About Novee:Join us in building Novee, the AI platform transforming how insurance risks are assessed, backed by $2.2 million in funding from 42Cap and Insurtech Gateway. We’re already working with some of the largest insurers, processing hundreds of millions in premium. Commercial & Specialty insurance is a $1 trillion industry that quietly powers the global economy. Behind every plane, power plant or stadium is a complex web of risk decisions that still rely on manual processes and gut feel. Underwriters trawl through documents and emails just to decide whether to take on a risk. We turn that clutter into fast, structured insights so insurers can make better, faster decisions. If you want to help transform a critical, overlooked industry with real-world impact, we’d love to talk.Postion Overview:We’re looking for someone to sit at the intersection of product, AI / LLM engineering, and data operations. You’ll be responsible for shaping how our AI learns and improving the accuracy, reliability and intelligence of the system that powers our AI underwriting platform.This is a highly impactful and hands-on role for someone who blends analytical capability with operational discipline. You’ll design the standards that define “high quality AI,” manage the process and creation of datasets that train and evaluate our models, investigate why AI gets things wrong, and work directly with engineering to drive measurable improvements. Your work will directly impact model accuracy, client outcomes, and product success.Key Responsibilities:Ground Truth & Data Labelling Operations:Own the creation and ongoing improvement of high-quality ground-truth datasets used for training and evaluationDevelop and refine labelling guidelines to ensure consistency and clarity for offshore labelling teamsManage and audit labelled data to ensure quality, reliability, and throughput targets are achievedAI / LLM Model Investigation & ImprovementInvestigate inaccurate model outputs to identify root causes and recurring patternsDevelop hypotheses around failure modes (e.g., document structure, formatting issues, ambiguous wording, model misinterpretation)Work directly with engineering to design experiments, validate improvements, and measure impact before releaseCross-Functional CollaborationPartner with product management to ensure evaluation priorities align with business and client needsWork closely with engineers to influence technical direction, testing strategy, and model developmentProvide clear insight and structured reporting to internal and external stakeholders on AI performance and reliabilityTesting & Release ValidationDevelop and execute test plans for model updates and new capabilitiesSign off release readiness based on objective performance standards and acceptance criteriaCommunicate improvements and accuracy changes clearly with relevant teams and stakeholdersAbout You:2+ years experience working in commercial or specialty insurance (e.g., Underwriting Analyst, Technician, Ops Specialist, Exposure Analyst, Wordings Analyst, etc.)Strong analytical and problem-solving skills with exceptional attention to detailComfortable interrogating data, spotting patterns, and making decisions based on evidenceOperates well in fast-moving environments with evolving priorities and high expectationsCurious, resourceful and motivated by building new capability, not maintaining status quo