endorsedOngoing

LLM-powered OCR and Document Classification for Life & Health Insurer

Services Covered:

AI and Machine Learning

Industry

Financial Services

Duration

17 months

Budget

200K - 999K

Start Date

1 March 2025

Team Size

1-5 Employees

Engagement Model

Onshore

Project Capability Score (PCS)

76/100

Good

Project Capability Score (PCS) estimates how capable BlueLabel is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

Assurity Group, Inc.

Lincoln, United States of America

Project Summary

BlueLabel provided AI development services to a life and health insurance carrier, building an LLM-assisted OCR and document classification solution to support claims adjudication. The team delivered a working proof of concept within days and proceeded to a full engagement that included integration to match extracted content with policy data. The solution improved efficiency, reducing task times from roughly 40–45 minutes to 10–12 minutes, and BlueLabel met deadlines and stayed on budget while offering post-implementation support.

Key Challenges

Handle increased claims adjudication volume without adding staff
Implement AI integrations for matching without in-house expertise

Project Deliverables

Working proof of concept
LLM-assisted OCR and image/text extraction
Document classification system
Policy-matching integration
Post-implementation support, monitoring, and optimization

Project Solution

BlueLabel developed an AI solution beginning with a rapid proof of concept and then moved to a full engagement. They implemented LLM-driven OCR and image/text extraction using Gemini and Llama Cloud as the LLM engines to perform document classification and to match extracted information against the client's policy data. The engagement included integration work, and BlueLabel provided post-implementation support, monitoring, and optimization throughout the engagement. The vendor staffed the project with two developers, a project manager, a business analyst, and a solutions architect.

Project Outcome

Tasks reduced from 40–45 minutes to 10–12 minutes
Some manual tasks eliminated through automation
Project met deadlines and stayed on budget with thrice-weekly stand-ups
Vendor rapidly learned domain knowledge and partnered on the best solution

Platforms

AI/ML Platform
Cloud
API/Integration Platform

Tech Stack

Google Cloud Platform (GCP)

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“The solution has improved efficiency and automation, reducing the time spent on tasks from 40–45 minutes to 10–12 minutes. BlueLabel meets deadlines and stays on budget, holding regular stand-up meetings to ensure smooth progress. Their ability to learn and apply newer technologies is impressive.”
Eric Otterstein
SVP & CIO
Date: 08/28/2025