[ Industry case study — Custom Software ]

InsurTech unicorn rebuilds its employee platform and claims operations

STX Next delivered a new agent platform, real-time automated claims processing and a Snowflake data platform for UK InsurTech unicorn Zego.

Original project by STX Next ↗Industry reference — not a NexSum Labs engagement.

Client
Zego
Industry
Insurance / InsurTech
Services
Software engineering, Data engineering, Systems integration, Legacy modernization
Year
2026
Location
United Kingdom
STX Next and Zego case study banner with a smiling driver behind the wheel
575,000+
drivers trusting Zego
3
production systems delivered
~100
structured Snowflake analytical tables

The client

Founded in 2016, Zego is a UK leader in flexible commercial motor insurance for rideshare drivers, couriers and fleet operators, and has reached unicorn status. It is trusted by more than 575,000 drivers and has recently expanded into credit products.

The challenge

The internal tool used by customer agents was already on its third iteration and still lacked the design quality and extensibility the business needed. Millions of policy, customer and vehicle records sat in inconsistent databases inside a legacy monolith, making reporting unreliable. Claims handling depended on manual Salesforce workflows that made real-time policy validation and automated fraud checks difficult as volumes grew.

The solution

01New employee platform

A clean-slate React.js frontend on top of extended Python/FastAPI services with gRPC and protobuf, including a custom policy cancellation engine and a live link to the UK government's claims management system.

02Automated claims processing

Python and Django microservices validate policies the moment an incident is reported, trigger fraud detection and integrate ShareDo, Stripe, Twilio and Zapier.

03Snowflake data platform

Apache Airflow pipelines on AWS migrated millions of legacy records into roughly 100 structured Snowflake tables that feed dashboards and downstream apps.

04AI-augmented development

Cursor AI and Claude Code were introduced with guardrails so engineers, and even some non-technical staff, can safely extend the platform.

The results

All three workstreams are live in production. Agents embraced the new portal, and Zego's new credit product launched with its agent workflow built in from day one. Claims are validated and screened for fraud in real time, manual data-entry errors dropped, and a centralized warehouse replaced fragmented databases. The partnership continues with deeper AI features and further legacy migration.

Tools & stack

PythonDjangoFastAPIReact.jsNode.jsgRPCprotobufPostgreSQLApache AirflowSnowflakeAWSShareDoStripeTwilioZapierCursor AIClaude Code

Original project by STX Next ↗ — published by the delivering agency, included here as an industry reference.

Next case study — Vestmark

Rails consolidation delivers 50% faster releases for Vestmark