Dvara is a fintech company that provides loans to a wide customer base. Historically, all of Dvara's finance-related BI reports were generated by querying data directly from on-premise production databases.

Business Challenges

As transaction volumes grew, querying production directly for reporting created both performance risk and data governance gaps.

Production performance risk

Growing transaction volumes meant reporting queries against production databases caused significant performance slowdowns.

External client exposure

Production-linked reports were shared directly with external clients, exposing live operational systems to reporting load and dependency.

Fragmented source systems

Finance data was scattered across six different systems and database types, from MySQL and MongoDB to SQL Server.

No self-serve access

Clients had no dedicated way to access their reports without depending on production systems directly.

Project Objectives

Geopits designed and implemented a cloud-based data warehouse on GCP to consolidate Dvara's data, decouple reporting from production, and deliver reports through a dedicated BI layer.

Key Goals:

Decouple BI reporting from production databases
Consolidate six source systems into one governed warehouse
Give clients self-serve access to reports without touching production

Solution Provided by Geopits

Geopits built an end-to-end pipeline on Google Cloud, from real-time change data capture through to an AI-assisted BI layer, replacing the old production-linked reporting model entirely.

Real-time CDC ingestion

Google Cloud Datastream captures changes from each source system and lands them in a GCS bucket in Avro format, forming the raw Bronze Layer.

Event-driven Silver Layer

Pub/Sub and Dataflow convert raw Avro files into SQL-compatible format, loading a clean, query-ready Silver Layer in Cloud SQL.

Modeled Gold Layer

A scheduled Cloud Run transformation service builds fact and dimension tables in the Gold Layer, the analytics-ready warehouse.

AI-powered BI and reporting

Vanna AI enables natural language querying on the Gold Layer, while Superset and a new Android app deliver reports directly to clients.

Results & Business Impact

Moving reporting off production and onto a governed warehouse changed how Dvara's teams and clients access data.

Capabilities

Before

After

Reporting source

Queried production databases directly

Queried from a dedicated Gold Layer warehouse

Data sources

6 separate, disconnected systems

Consolidated into one governed warehouse

Client report access

Shared directly, tied to production

Self-serve via a new Android app

Querying

Required SQL knowledge

Natural language querying via Vanna AI

Key outcomes

Reporting decoupled from production

Finance and business reports now run off the Gold Layer, removing reporting load and dependency from production systems.

Unified data foundation

Six previously separate source systems, spanning MySQL, MongoDB, and SQL Server, are consolidated into one governed warehouse.

Self-serve client access

A new Android app gives Dvara's clients direct, self-serve access to their reports without touching production.

AI-assisted querying

Vanna AI lets business users query the warehouse in natural language, without writing SQL directly.

Conclusion

Geopits partnered with Dvara to build a modern, cloud-native data warehouse on GCP, decoupling BI reporting from production, consolidating six source systems, and delivering reports through an AI-assisted BI layer and a new client-facing Android app.

Ready to Transform Your Data?

Geopits works alongside your team as a strategic partner, starting with stabilizing your current databases, then modernizing your data infrastructure, and ultimately helping you unlock the full potential of AI.

Book a Demo Learn More

170

Happy Clients so far

2100+

Databases Managed

142+

Successful Migrations

At Geopits, our mission is to empower businesses to achieve reliable, secure and high-performing database environments and to derive meaningful business insights.