FinTechCase study

Personal finance, automated end to end.

Pennygood turns raw transaction data into automated, personal financial guidance — a consumer FinTech platform Codeiqra designed, built and still operates.

Total balance

$2,481,920

▲ 18.2% this quarter

Payments

automated

Compliance

automated

Payroll

automated

Real-time

Insight latency

Automated

Fraud & anomaly screening

Event-driven

Core architecture

Zero-downtime

Migration

The challenge

Growth was outrunning the stack.

Pennygood had product-market fit and a monolith that couldn't keep up: nightly batch insights, manual categorisation reviews, and an infrastructure bill growing faster than the user base. Every new feature made the next one slower to ship.

What we built

An AI-first platform rebuild — without stopping the business.

We re-architected Pennygood as an event-driven platform with ML at the core, migrating from the monolith by strangler pattern while the product kept shipping. Insights moved from nightly batches to real time; categorisation and fraud checks became fully automated with human review only on low-confidence cases.

▸Real-time transaction categorisation (ML)

▸Personalised insight & nudge engine

▸Event-driven core on Kafka

▸Automated fraud & anomaly screening

▸Zero-downtime strangler migration

▸CI/CD with progressive rollouts

The outcome

Category-leading engagement, boring operations.

Engagement moved measurably ahead of category benchmarks, insight latency dropped from hours to real time, and the automated pipeline now handles transaction volume at scale with a smaller ops team than before the rebuild. The platform has run reliably ever since.

At a glance

Industry

Consumer FinTech

Engagement

Dedicated team · Ongoing

Scope

Platform rebuild + AI automation

Team

Embedded pod

Status

Live · under managed SLA

React NativeNode.jsKafkaPostgreSQLPyTorchAWS

The kind of feedback we build toward

“We've bought a lot of software that got a launch email and then nothing. Codeiqra built the first thing in years my whole team uses without being told to.”

Engineering Lead

Pennygood

Next case study

Amperly — EV Charging Network

Read it →

Scaling a consumer platform of your own?

We'll walk you through the Pennygood architecture and what it would look like applied to your product — candidly, within two business days.