AI-powered SaaS platform
for commercial insurance
analysis
Built a commercial insurance SaaS platform that brings policy documents and structured insurance data into one searchable environment. The platform supports document processing, policy comparison, and AI-assisted coverage review.
Business context
CoverageXpert is an insurance intelligence platform that turns domain expertise and complex policy documents into a SaaS product for commercial insurance review. It enables insurance professionals to search policy information, reference source documents, identify discrepancies, and generate visual change reports.
Project in facts
- Insurance
- USA
- 2 Ruby on Rails developers, UX/UI designer, Business analyst, PM
- April 2020 – March 2026
- MVP development, Ruby on Rails development, UX/UI design, AI integration
- Ruby on Rails, PostgreSQL, Sidekiq, AWS, Google OCR
Challenge
The client needed a centralized SaaS platform to streamline access to complex commercial underwriting data. Their objective was to aggregate information from multiple legacy sources into a unified database and equip users with tools to efficiently search, compare, and analyze coverage details.
Product goals
- Centralize commercial insurance data in a searchable knowledge base
- Enable fast and accurate comparison of policies and documents
- Help brokers make faster and more informed coverage decisions
Solutions
We developed two connected modules: a SaaS application for searching and comparing documents, and a data-processing pipeline that normalizes data from multiple sources, stores it in a structured database, and links it to the corresponding policy documents. This created a unified data layer where users can search, compare, and analyze structured insurance information alongside source documents.
The platform equips brokers with a single workspace to search policy forms, review endorsements, and access related documents. To support the business model, we deployed a freemium monetization strategy: basic database browsing is accessible at no cost, while policy document retrieval and advanced comparison analytics operate on a usage-based model.
Following the MVP launch, we integrated AI-assisted insurance analysis, enabling users to query the database and receive answers with references to the original documents they are authorized to access.
Insurance database
Built a searchable database with document viewing and policy comparison tools.
Document processing
Developed 12 custom parsers and an OCR pipeline to extract structured data from PDFs.
AI policy review
Integrated ChatGPT and Gemini to help users query the database and explore coverage-related information.
SaaS infrastructure
Implemented authentication, Stripe payments, store credits, and a pay-per-document model.
Engineering approach
Leveraging Ruby on Rails development, we built a platform with PostgreSQL to manage complex relational data across policies, documents, and endorsements. To support the client’s document-heavy workflows, we engineered an ingestion pipeline combining Google OCR with 12 custom parsers, automatically converting raw PDFs into a searchable knowledge base.
A key technical challenge was OCR performance: standard processing exceeded one hour. We split documents into individual pages and processed them in parallel, reducing processing time to approximately 5–10 minutes.
To expand the platform’s core capabilities, we integrated Stripe, SendGrid, and Gemini. Sidekiq handles external API calls and heavy parsing workloads through background queues, keeping the application responsive.
Business impact
Result
Rubyroid Labs delivered a SaaS platform for commercial insurance policy analysis.
-
Enterprise-ready product
Brokers can search, compare, and analyze commercial insurance products in a centralized workspace. -
Faster document processing
Parallel processing cuts document processing time from over one hour to 5–10 minutes.
A skilled and meticulous partner, Rubyroid Labs delivered a product that exceeded expectations. Their project management skills and clear communication made for a collaborative process. In addition to being responsive, they made a recommendation that ultimately led to the project’s success.
Rick Lusk, Plaus Tech 
Build a SaaS around your industry expertise
Please fill out this form, and our manager will contact you within one business hour. If necessary, we can sign an NDA and begin project discussions.
Thank you.
Your message has been sent successfully!
We’ll get in touch with you within 24 hours, excepting requests received on Saturday, Sunday.
