Building a Real-Time,
Data Mesh Platform for Insurance
Designed and built a production-grade data mesh supporting real-time and batch workloads across under strict regulatory and operational constraints.
The Client
Global specialty insurance group operating across multiple markets and lines of business, with a strong reliance on data for underwriting, claims, risk management, and regulatory reporting. The organisation runs complex, high-volume data workloads under strict governance, security, and compliance constraints.
4
0
9
4
3
7
8
6
4
1
0
4
3
2
7
8
0
4
2
0
+
Countries of operation
4
0
9
4
3
7
8
6
4
2
0
4
3
2
7
8
0
4
2
0
+
Line of businesses
4
0
9
4
3
7
8
6
4
1
0
4
3
2
7
8
0
4
2
0
0
4
3
2
7
8
0
4
2
0
0
4
3
2
7
8
0
4
2
0
+
Employees worldwide
The Challenge
The organisation had an established on-premise data architecture supporting a wide range of production use cases, but the platform was not designed for the scale, speed, and delivery model now required. Modernisation initiatives had started, yet progress was slow due to architectural complexity, legacy constraints, and the need to protect critical workloads already running in production.
At the same time, the business required near real-time data to improve underwriting, claims, and risk insights, while enabling data science teams to work with fresher, more reliable information. This had to be achieved in a highly regulated environment, where data governance, security, lineage, and access control were mandatory and could not be treated as afterthoughts.
The challenge was therefore to evolve the existing platform into a modern, real-time, cloud-ready data platform, improving insight and delivery speed, while strengthening data governance and maintaining full operational and regulatory compliance.
At the same time, the business required near real-time data to improve underwriting, claims, and risk insights, while enabling data science teams to work with fresher, more reliable information. This had to be achieved in a highly regulated environment, where data governance, security, lineage, and access control were mandatory and could not be treated as afterthoughts.
The challenge was therefore to evolve the existing platform into a modern, real-time, cloud-ready data platform, improving insight and delivery speed, while strengthening data governance and maintaining full operational and regulatory compliance.
Our Solution
A global Data Mesh Architecture providing governed real time and analytical data products to the business
Domain driven Architecture
Designed and implemented a production data mesh structured across five core domains: Underwriting, Rating, Broker, Cyber, and Central. Each domain owns its data products while relying on shared platform services for governance, security, and operational consistency.
Real-Time Data Products
Built high-volume streaming data products using Confluent Kafka, enabling near real-time processing of operational events across submission, policy, claim, premium, and cyber workloads. Designed for scalability, resilience, and low-latency delivery.
Analytical Data Products
Delivered governed batch data products in Snowflake and ADLS, enabling scalable analytics, regulatory reporting, and complex financial and actuarial processing workloads.
Multi-Regional Azure Platform
Engineered a multi-region Azure platform to support distributed teams, regulatory boundaries, disaster recovery, and high availability—ensuring production resilience across markets.
Core Insurance Data Products
Implemented domain-level data products including Submission, Policy, Claim, Premium, Workbench, and PMDR, supporting underwriting, regulatory reporting, and operational analytics in production.
Data Catalog (Data.world)
Integrated Data.world as the enterprise data catalog to manage domains, data products, lineage, and ownership. Extended the catalog to integrate directly with Kafka topics, bridging streaming metadata with governed domain models.
The Value
The Modern Data Platform transformed how business teams access, trust, and act on data—enabling faster decisions, stronger governance, and scalable growth.
4
0
9
4
3
7
8
6
4
0
5
4
3
2
7
8
0
4
2
0
.
Data domains onboarded
Core domains onboarded with the platform launch: Underwariting, Claim, Broker, Finance, Central
4
0
9
4
3
7
8
6
4
5
0
4
3
2
7
8
0
4
2
0
+
Engineers working on the Platform
Multiple teams working in parallel for the different domains following identical delivery practices and standards
4
0
9
4
3
7
8
6
4
1
0
4
3
2
7
8
0
4
2
0
+
Real Time Data Products
Core Data Products (Policy, PolicySection, Claim, etc.) and Derived DPs (Premium, PMDR) generated in Real Time.
.svg){kind=icon}
Faster, Better Underwriting & Risk Decisions
Real-time streaming data products make submissions, policies, claims, and cyber events available almost immediately. This gives underwriting and risk teams faster access to data, improving responsiveness and enabling more timely, informed decisions across core insurance processes.
Trusted, Governed and Compliant Data
An integrated Data Catalog provides visibility of domains and data products, improving discovery, transparency, and controlled access. In a regulated insurance environment, this ensures compliance and audit readiness—while giving business teams confidence in the data they use.
.svg){kind=icon}
A Platform That Scales With the Business
A modular, domain-driven architecture enables federated delivery of data products, allowing teams to work in parallel under shared standards. This increases productivity and supports growth across new domains, regulatory requirements, and future data and AI initiatives.
Talk Less. Build More.
© 2026 All rights reserved