How Empyrean Brings Client Trends to the Table in 30 Seconds With Level AI Workers
Empyrean Benefits Solutions runs a service center that helps employees navigate medical, dental, vision, supplementary, and financial benefits. The center handles hundreds of thousands of calls a month. Each one a moment that matters for a member trying to make a decision for their family.
Freddie Berberena describes AI Workers in unusually direct terms:
AI Workers is the coolest thing that you've launched yet.
This is a short story about why.
The Problem: Weekly Client Meetings Without Fresh Trend Evidence at Hand
Empyrean meets with employer-group clients on a weekly basis. Those meetings are the moment when service center insight either becomes useful to a client or stays trapped inside the contact center.
Before AI Workers, surfacing trends from across thousands of conversations required time, skill, and analytic effort. Empyrean had already moved from reactive to proactive servicing as part of their broader Level AI deployment, and the weekly client meeting was the place where that posture still needed sharper tools. Walking into a meeting with a current read on what was happening across a client's member population meant either preparing in advance or going without the evidence.
The Solution: A Prompt-Driven Workflow That Pulls Trends in 30 Seconds
AI Workers gave Empyrean's leaders a way to ask their conversation data a question and get a well researched and reasoned answer back in the time it takes to write the prompt. The same interface serves two distinct audiences inside the business.
Having that level of information at your fingertips and being able to bring that to your client or bring it back internal to do coaching has been really cool.
Two use cases sit inside that quote. The Conversation Research Worker pulls trend and data evidence for client-facing conversations. The Coaching Plan Worker creates coaching plans backed by evidence for every agent.
What 30 Seconds Changes About the Client Conversation
The 30-second figure is not a statistic about average prompt latency. It is Freddie's description of the operational rhythm AI Workers creates inside his team.
In a weekly client meeting, a service center leader who can pull a real trend on the spot becomes a different kind of partner to their client. The conversation moves from "let me get back to you with that" to "here is what we are seeing across your population."
It is like that first step that's been made available to us where we can be proactive in our conversations and communications with our clients that we have meetings with on a weekly basis. And it takes us 30 seconds to pull that information out.
The compression of time between question and answer is the mechanism. The change in the client conversation is the outcome.
What Empyrean Actually Asks AI Workers Every Day
The breadth of use cases inside Empyrean's AI Workers deployment is wider than the two Freddie mentioned. The team prompts AI Workers for client-ready trend reports across dozens of employer-group campaigns, building per-campaign briefings on top call drivers, subtopics, and week-over-week shifts that feed directly into client conversations. They also run pod-level daily manager briefings, where the Team Performance Worker produces a structured executive digest of yesterday's top issues, affected clients, and emerging trends for each pod of agents.
Empyrean also uses AI Workers as a compliance and risk surveillance tool, prompting for supervisor-request patterns, possible call-avoidance behavior, incidents where wrong information may have been shared with members, and repeat-caller root cause analysis across 60-day windows. The common thread across every prompt is that a question that would have taken an analyst hours to answer becomes a 30-second query against the existing conversation dataset.
Freddie is candid that Empyrean is early in its AI Workers rollout. The team is still building internal guardrails and still mapping what the tool can do.
I'm sure we don't know half of what the power of that tool is yet because it is in a new state, but we're really excited to grow and take advantage of that and can see a whole ton of value in what it's gonna bring for us.
This is the honest version of an AI Workers adoption story. The 30-second prompt is already changing client conversations, and Empyrean is still uncovering the rest of its potential.
Why This Matters: Conversation Data Becomes Conversational on Demand
Empyrean's experience points to a specific capability AI Workers bring to a service organization. A service center that has spent years gathering conversation data can now make that data conversational on demand. A leader walking into a client meeting does not need an analyst to prepare a deck. The leader writes a prompt, gets the trend, and brings it to the table.
For an industry where service centers were historically reactive, that change is the operational shift. Freddie's team sees the early outcome already. The rest of what AI Workers can do for them is still being mapped.
See what AI Workers can do with your own conversation data. Request a demo
Keep reading
View all
