Insurance is one of the most operationally intensive industries in the world. Behind every policy is a mountain of work - submissions to triage, renewals to process, claims to investigate - and much of it still moves forward by hand, one document, phone call, and system update at a time.

For decades, automating insurance work meant a tool for each step. OCR digitized documents, RPA wrote results back into policy systems, and IVRs routed calls. Each task still required human handoffs to move from intake to outcome.

AI agents can now complete the whole task. Today, we're launching one multimodal agent that runs every step, end to end. It picks up the phone with the broker, reads the loss run, reasons over your procedures, operates your policy system, and closes the file. One agent for the entire unit of work.

Working across every modality

Three core capabilities enable our agents to run hundreds of thousands of tasks at best-in-class SLAs:

1. Procedures
   
   The Agent Operating Procedure (AOP) is our single surface area for orchestrating the work. Traditionally, operations tasks have been codified as Standard Operating Procedures (SOPs) spanning tens of pages of instructions for operations teams to complete hundreds of steps. With AOPs, you write the same steps in plain English, but then the agent carries them out.
   
   

2. Subagents
   
   Accuracy comes from focus, so Pace agents work several channels at once by splitting the AOP into focused pieces. The managing agent delegates work to subagents. Each are experts in their specific task: a voice agent for the phone call, a computer-use agent for navigating the admin system, a document agent for reasoning over a policy document. As a result, you get specialist’s accuracy, while the manager ensures close adherence to the AOP.
   
   

3. Shared context
   
   To make sure nothing gets dropped as a case moves, we give every agent one continuous thread of context that follows the work across every channel. The detail it pulled from the customer’s FNOL is still in hand whether it's on the phone an hour later for missing information or entering the detail back into the admin system. As a result, our agent can carry a case from start to finish, instead of losing context each time the work changes hands.

A demo of a Pace agent autonomously completing a premium audit across voice, email, document extraction, and computer use

Building trust for mission-critical, regulated work

Getting from demo to production takes more than just a capable agent. When a single answer is stitched together from documents, systems, and calls, our regulated customers require audit trails down to the keystroke. Pace agents show their work, in a few ways.

1. Citations
   
   Pace agents cite and log every decision they make. Take any figure and you can follow it back to the exact source behind it: the page in a loss run it was read from, the line in a call transcript where it was confirmed, the field in your policy system it was written to. Human handoffs used to make this granular observability impossible. Now, each agent’s workspace is recorded so that every action the agent takes in your systems has a screen recording and screenshots to show its work.
   
   

2. Graders
   
   Where citations show the work, graders check it. Graders are automated quality checks against the agent’s work. You define what correct looks like, and the graders runs automatically on every case. Some graders read the agent's work and decide whether it followed your procedure, like confirming the policy number before quoting a renewal. Others are queries that check calculations, like whether the audited premium equals the sum of the premium per class code.
   
   

3. Testing
   
   Finally, test cases let you simulate the agent across many runs at once. A test case is a real example paired with the answer you already know is correct. Our customers run hundreds of test cases which becomes their evaluation benchmark for accuracy. Every run, and every correction your team makes, becomes another test the agent is measured against. This compound loop makes the agent more reliable on your work over time, similar to training a teammate.

A demo of Pace’s citations, graders, and test cases to ensure our multimodal agents stand up in production

In production today

For as long as anyone can remember, growth scaled linearly with headcount. Agentic operations has changed that. The expertise and judgment stay with your best teammates while their work is supercharged by a team of AI agents.

This future is happening now. Pace agents are running tens of thousands of these tasks a week at customers like Prudential, WTW, Palomar, and Convex. If you run an insurer and share our vision, let’s build the future together.