Posts on what we actually think.
Essays on conformal prediction, social inflation, and where generative AI helps and where it doesn't. Written by the team. Slowly.
Showing 9 posts · Product · Tal Knafo
Clear filtersLatency vs Depth: Engineering Real-Time Case Analysis
Processing thousands of unstructured claims documents requires an architectural choice between speed and depth. Building a system that actually informs reserves means accepting that text extraction is slow and math is fast.
Calibration Monitoring in Production
A single point prediction for an insurance settlement is a lie. We monitor the boundaries of our predictions, ensuring a 30 percent probability of nuclear escalation means exactly 30 out of 100 cases will explode.
Fitting Into an Existing Claims Workflow
Claims adjusters live in their core systems. Forecasting litigation risk requires an architecture that operates entirely in the background and delivers calibrated answers exactly where the work actually happens.
Reserve Delta, Escalation Probability, Comparables: The Feature Set
A single point prediction is useless in litigation. Claims forecasting requires calibrated ranges and traceable drivers to survive the scrutiny of a reserving committee.
Want to talk to an executive?
Press, partners, investors, candidates — the inbox is monitored. Tell us who you are and we'll route it to the right person within two business days.