Data Scientist, Agentic AI (Insurance Underwriting) (GL89)

Data Scientist, Agentic AI

We're on a transformation journey to evolve into a modern, forward-thinking insurance company committed to enhancing the wellbeing of its customers and their families. This role presents a distinctive opportunity to drive real-world impact by applying cutting-edge AI to transform how we do business.

Our Data & AI team spearheads a culture of intelligence and automation across the enterprise, creating business value from advanced data and AI solutions. Our team includes data scientists, engineers, analysts, and product leaders working together to deliver AI-driven products that power growth, improve risk management, and elevate customer experience.

We created the Data Science Lab (DSL) to reimagine insurance in light of emerging technology, evolving consumer needs, and rapid advances in AI. The DSL expedites our transition to data-driven decision making and fosters innovation by rapidly testing, scaling, and operationalizing state-of-the-art AI.

We are seeking a Data Scientist, Agentic AI—an experienced individual contributor with strong experience in Agentic AI, large language models (LLMs), and natural language processing (NLP) and a track record of turning advanced research into practical, impactful enterprise solutions. This role focuses on building, deploying, and scaling agentic AI systems, large language models, and intelligent automation solutions that reshape how Guardian operates, serves customers, and drives growth. You'll collaborate directly with senior executives on high-visibility projects to bring next-generation AI to life across Guardian's products and services.

Key Responsibilities:

• Design and implement Agentic AI solutions that automate business workflows, improve decision-making, and enhance customer and employee experiences.
• Apply LLMs and generative AI to process and interpret unstructured data such as contracts, underwriting notes, claims, medical records, and customer interactions.
• Develop autonomous agents and reasoning systems that integrate with Guardian's platforms to deliver measurable business outcomes.
• Collaborate with data engineers and AIOps teams to ensure models are scalable, robust, and production-ready.
• Translate research in agentic AI and reinforcement learning into practical applications for underwriting, claims automation, customer servicing, and risk assessment.
• Work closely with product owners, engineers, and business stakeholders to define use cases, design solutions, and measure impact.
• Contribute to the Data Science Lab by building reusable components and frameworks for developing and deploying agentic AI solutions.
• Adhere to AI and LLM governance, documentation, testing, and other best practices in partnership with key stakeholders.

You are:

• Passionate about applying advanced AI techniques to solve real-world business challenges.
• Curious about agentic AI, autonomous systems, and LLM-based solutions that transform industries.
• A hands-on builder who enjoys moving solutions from prototype to deployment.
• Comfortable collaborating in cross-functional teams and aligning technical solutions with business goals.

You have:

• PhD with 0–1 years of experience, Master's degree with 2+ years, or Bachelor's degree with 4+ years in Statistics, Computer Science, Engineering, Applied Mathematics, or related field.
• Experience in insurance industry - mandatory
• Underwriting Experience is Preferred
• 2+ years of hands-on experience in AI/ML modeling and development.
• Solid understanding of probability, statistics, and machine learning fundamentals.
• Strong programming skills in Python and familiarity with frameworks like PyTorch, TensorFlow, and LangGraph.
• Experience with LLMs, generative AI, and multi-step reasoning systems.
• Excellent problem-solving and analytical skills with attention to detail.
• Strong communication skills and ability to collaborate effectively with product and engineering teams.
• Working knowledge of core software engineering concepts (version control with Git/GitHub, testing, logging,...).
• Working knowledge of a variety of machine learning techniques (clustering, decision tree, bagging/boosting artificial neural networks, etc.) and their real-world advantages/drawbacks.

Location:

• Three days a week at a Guardian office in New York, NY, Holmdel, NJ, Bethlehem, PA, or Boston, MA