Systems architect São Paulo, Brazil

Fernando
Gomes

20 years building critical systems in payments, banking and healthcare.

Architecture. Engineering. Technical leadership.

From the first decision to production.

Selected work

Systems built for the demands of real operations.

Healthcare

A common language for clinical data.

A health data platform built around FHIR, giving clinical systems and external partners a consistent way to exchange information.

Data architecture · Interoperability

50%less time to integrate partners

Financial systems

Modernizing the core.

Calculation, accounting and contract engines for regulated banking operations. A microservices practice to guide the move from monoliths to distributed systems.

Core platforms · Modernization

R$10B+in financial assets supported

Applied AI

From two days to thirty minutes.

AI-agent workflows that read, check and assemble accountability reports, reducing the manual work in a demanding reporting process.

Agent workflows · Process automation

30 minfor a process that took two days

Public infrastructure

Planning for national scale.

A technology roadmap and cloud architecture for a national transaction platform, balancing modernization with the demands of an ongoing operation.

Cloud architecture · Technology strategy

Billionsof annual transactions planned for

The decisions
behind the system.

I work across architecture, hands-on engineering and technical leadership. My experience spans financial platforms, healthcare data, cloud infrastructure and applied AI.

Every design choice has a consequence: availability, cost, regulatory risk or the time it takes a team to deliver.

Download résumé

Start with the problem.

Before choosing a technology, understand what needs to change and how to tell if it worked.

Modernize with context.

Existing systems carry years of business knowledge. Moving forward means understanding what to keep.

Build AI with boundaries.

Useful AI needs a defined task, reliable inputs and a way to verify the result.

Writing

Notes on software and the choices behind it.

All articles

Beyond the Chatbot: Building Agentic Workflows for Real-World Due Diligence

Why 'Chat with PDF' isn't enough for business strategy, and how we engineered Python-based AI Agents to automate complex investment analysis.

From Monoliths to Microservices: What 15 Years of Java Taught Me About Node.js

Why a veteran Java Architect embraced Node.js for Cloud-Native systems, and how to apply Enterprise rigor to modern JavaScript runtimes.

Get in touch

Have a hard problem
to work through?

hi@fernandogomes.dev