Enosis Outsourcing logo
EndorsedendorsedCompleted

RelmAI — Azure-based AI Search & Underwriting Chat

Services Covered:

AI and Machine Learning • Custom Software Development • UI/UX Design
Project hero — replace with case study imagery
Industry
Industry

Information Technology

Duration
Duration

6 months

Budget
Budget

200K - 999K

Start Date
Start Date

1 February 2024

Team Size
Team Size

1-5

Engagement Model
Engagement Model

Offshore

Project Capability Score (PCS)

gauge meter
81/100
Strong

Project Capability Score (PCS) estimates how capable Intersog 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

Freestone AI

map Bermuda

Project Summary

INTERSOG built RelmAI, an Azure-based platform that transforms large volumes of unstructured insurance data into structured insights and provides an LLM-driven chat interface for underwriters. The solution included data ingestion from SharePoint, indexing and enrichment pipelines, PostgreSQL storage, and a multi-layered security architecture, resulting in faster, more reliable underwriting workflows.

Key Challenges

  • Identify an AI solution to increase the speed of underwriting
  • Build data pipelines to feed LLMs and create structured output from a large unstructured document store
  • Create an LLM chatbot and UI/UX for the underwriting team to engage with

Project Deliverables

  • Azure-based RelmAI platform
  • AI-powered search engine with generative AI chat interface
  • Data ingestion and indexing pipelines (SharePoint ingestion, AI Search indexer)
  • Data enrichment via Azure Functions (classification, tagging, ChatGPT-generated summaries)
  • PostgreSQL database for structured storage
  • Multi-layered security architecture using Microsoft Entra ID

Project Solution

Together with Freestone.AI, INTERSOG developed RelmAI, an Azure-based AI platform that indexes and enriches documents from SharePoint, builds structured data in PostgreSQL, and exposes a generative AI chat interface for underwriters. The vendor implemented data pipelines and custom Azure Functions for classification, tagging, and content summarization, integrated off‑the‑shelf LLM capabilities, and deployed a multi-layered security architecture using Microsoft Entra ID.

Project Outcome

  • Transformed large volumes of unstructured documents into structured, queryable insights
  • Enriched data with contextual metadata (classifications, tags, summaries) stored in PostgreSQL
  • Delivered an LLM-driven conversational interface that improved underwriting speed and quality
  • Implemented enhanced search with source-document verification and a multi-layered security architecture

Platforms

  • CloudCloud

Tech Stack

  • AzureAzure
  • PostgreSQLPostgreSQL

Client Endorsement

Overall Review Rating

5star

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“INTERSOG successfully transformed the client's unstructured data into structured insights, enabling a comprehensive view of client data. The team enriched the data by adding valuable context and implemented a multi-layered security architecture, much to the client's delight.”

Matt Curcio

CEO

Note: This endorsement is based on publicly available client feedback from external review sources.

More Projects by Intersog