About Northline AI
Northline builds extraction, classification and agent systems for finance, operations and support teams. We publish our numbers, including cost, exception rates and what we would do differently.
Business
Document and support automation for mid-sized companies, built on measured results.
On file
Is this your company?
Claim this page by verifying your website.
Claim this businessNorthline builds extraction, classification and agent systems for finance, operations and support teams. We publish our numbers, including cost, exception rates and what we would do differently.
Articles
Prompt caching, request batching and a routing layer did more than any model swap. Here is what we measured, in order.
Our first two attempts at automatic ticket routing made things worse. The third one shipped because we stopped optimizing for accuracy.
Six sentences we make every customer read and confirm before we enable autonomous replies. Most of the value is in what we exclude.
Most of our extraction tasks have no single correct answer to grade against. This is the scoring approach we landed on after two that failed.
Our agent could technically approve these on its own. The data on why we still do not is more interesting than the policy.