Current Focus

Data Engineering for Fintech

Currently running data and analytics at a UAE payments fintech. The data platform, real-time pipelines, and finance dashboards that sit downstream of the transactional systems.

Context

What data engineering looks like at a payments fintech

Data accuracy is non-negotiable. Every transaction ends up in a dashboard, a reconciliation report, or a finance KPI. The data platform has to be reliable under streaming load, correct across multiple entities and currencies, and understandable when someone from finance asks why a number moved.

My scope is the data platform itself: ingestion from internal systems and partner feeds, transformation, real-time pipelines, reconciliation data models, and the dashboards and alerting on top. The accounting logic, compliance program, and regulatory sign-off sit with the finance, risk, and security teams. I build what they need to do their jobs.

Capabilities

What I build for fintech

01

Reconciliation Data Models

Data models that let finance teams automate reconciliation work that used to live in spreadsheets. I build the pipelines and models. Finance defines the matching rules and signs off on the numbers.

  • Ingestion from internal databases and partner feeds
  • Normalized transaction models across sources
  • Exception surfaces finance can triage
  • Dashboards on top for day-to-day ops
02

Real-Time Transaction Pipelines

Streaming ingest of financial transactions with Spark Structured Streaming and CDC, feeding live dashboards and alerts so ops and finance see issues within minutes instead of the next morning.

  • Spark Structured Streaming with CDC from operational DBs
  • Multi-currency handling with live FX rates via API
  • Multi-entity transaction models
  • Operational dashboards and anomaly alerting

Platform Migrations

Moving off legacy ETL onto a modern platform. Delivered a Databricks migration that cut ETL cost by 50% while improving data freshness.

Finance & Ops Analytics

BI dashboards that translate transaction data into the KPIs finance, ops, and leadership actually use day to day.

Multi-Source Ingestion

Pulling together internal SQL and NoSQL databases, partner JSON feeds, and external APIs like FX rates into a single analytical layer.

Good Fit

Where this work applies

Payments and digital banking

Money transfer, wallets, payment processors, neobanks. My current engagement is in this space.

Other financial-data-heavy companies

E-commerce and marketplaces reconciling against processors, SaaS with complex billing. The data engineering work transfers even without domain specialization.

Results

Track record

50%

ETL cost reduction through a Databricks migration at a UAE payments fintech

Real-time

Streaming transaction pipelines with Spark Structured Streaming and CDC, feeding live dashboards and alerts

Finance-ready

Reconciliation data models that let the finance team automate work previously done in spreadsheets

Building out the data platform at a payments company?

Happy to talk through platform migrations, reconciliation models, or real-time pipelines.