Case study · LegalTech · AI

CadreAI

Multi-tenant SaaS that lets teams create, configure and deploy custom agents, knowledge bases and RAG flows for case automation.

Industry: LegalTech · AI
Relationship: Led by our co-founder at CadreODR
Live product: ai-service.cadreodr.com

Overview

CadreAI is the AI platform built inside CadreODR, a legal-technology company. It gives teams a self-serve way to build AI agents and knowledge bases for their own case workflows, instead of waiting on engineering for each new use case.

Challenge

Case work involves large volumes of documents and repetitive procedural steps. Different teams needed their own agents and their own knowledge — on a single platform, with data kept separate between tenants.

Approach

01
Designed as multi-tenant SaaS from day one, so each team's agents, knowledge bases and data stay isolated.
02
Made agents and RAG flows configurable by users, rather than hard-coding a fixed set of AI features.
03
Built the service layer in Python with FastAPI, deployable to GCP or AWS.

Architecture

Interface

Agent builder
Knowledge-base management
Flow configuration

Services

FastAPI APIs
Agent runtime
RAG pipeline

Data

Tenant-scoped knowledge bases
Document ingestion
Retrieval

Infrastructure

GCP / AWS

High-level view of the system's main layers.

What we built

Self-serve creation and configuration of custom AI agents
Knowledge bases built from each team's own documents
Configurable RAG flows for case automation
Multi-tenant platform with per-tenant isolation

Technology

Python
FastAPI
RAG
AI agents
Multi-tenant
GCP / AWS

Outcome

CadreAI is live at ai-service.cadreodr.com. A detailed write-up of results is being prepared. We're happy to walk you through the project on a call.

Screenshots

Screenshots for this case study are being prepared.

Next case studySajhaserve

Have a product like this to build?

Tell us what you're working on. We'll tell you honestly how we'd approach it.

Talk to an Engineer

Prefer email? info@vertexlab.com.np