Giving Homeowners Answers About Insurance Coverage in Seconds with Model-driven Policy Approvals

Topdanmark's staff manually reviewed building-condition reports for insurance claims and new policies, often taking nearly a week to give customers a coverage decision. It deployed ML models on Domino's platform (PyTorch, TensorFlow, Keras, Spark, Python, AWS) for automated policy approvals with API-based deployment and real-time drift monitoring. Policy approvals dropped from about 4 days to 1-2 seconds -- roughly 800x faster -- with 65% of cases automated (beating an initial 30-50% target) and minimum viable models built in 12 weeks. Head of Machine Learning Stig Pedersen said 'the model is 800 times faster than humans, so customers can learn whether they're covered in one to two seconds.'

Domino Data Lab is an enterprise MLOps platform built for regulated industries — the sanctioned sandbox where biopharma teams run reproducible experiments under GxP/21 CFR Part 11 guardrails.

This sits in the Design & Stress-Test stage of the AI use-case lifecycle: Named examples from the tools pharma teams use to scope, prototype and benchmark an AI use case before committing to build it — data-catalog, sandbox and evaluation platforms.