Software Engineer

Matheus Rodrigues

Backend, scalable APIs, observability, cloud infrastructure, and production reliability.

I build and improve systems that need reliable integrations, predictable behavior, and clear visibility in production.

Featured Technical Solutions

Production problems I like solving.

Load control, async processing, observability that actually helps, and fixes that make systems easier to trust in production.

Rate Limit and Queue with p-queue

Problem

When an API pushes an external integration too hard, rate limits hit quickly and failures start to spread.

Solution

I put the critical outbound flow behind a queue with defined concurrency and pacing.

Impact

The integration stayed within real capacity, with less noise and less cascading failure downstream.

Snippet

TypeScript / Node.js

const queue = new PQueue({
  concurrency: 4,
  intervalCap: 20,
  interval: 1000,
});

export function enqueue(job: InvoiceJob) {
  return queue.add(() => provider.send(job));
}

Scalable Async Worker

Problem

When heavy processing stays inside the API, latency grows and recovery gets harder.

Solution

I split ingestion from execution with a job queue and a dedicated worker.

Impact

The API stayed responsive, and the processing flow became easier to scale and rerun.

Snippet

TypeScript / Node.js

await jobs.publish("access-events", payload);

worker.start(async (job) => {
  await processAccessEvent(job.payload);
  await job.ack();
});

Production Observability

Problem

Without context in production, incidents turn into guesswork.

Solution

I standardized structured logs, request correlation, and latency tracking on the paths that matter.

Impact

It became easier to connect the incident to the right part of the flow and the real impact.

Snippet

TypeScript / Node.js

logger.info({
  requestId,
  route: req.path,
  latencyMs,
  statusCode: res.statusCode,
}, "request completed");

Reliability and Critical Fixes

Problem

Recurring critical issues usually point to weak contracts, incomplete validation, or missing idempotency.

Solution

I fixed the root cause and reinforced the flow with explicit validation and idempotent writes.

Impact

The system became more predictable and safer to operate.

Snippet

TypeScript / Node.js

await db.paymentEvent.upsert({
  where: { externalId: event.id },
  update: {},
  create: mapPaymentEvent(event),
});

Production Product

Production product, used in real operations every day.

QRFluid is a QR code-based access control and time tracking SaaS built for logistics operations, already running in production and processing more than 15,000 time records per month.

In production

Product

QRFluid

Built for operational routines that depend on fast registration, low-friction flows, and predictable day-to-day use.

Context

The core product is already running in production with QR code-based access control and time tracking, supporting routines where speed and reliability matter. Facial recognition and fingerprint biometrics are being implemented as the next product evolution, without changing the main flow already in use.

View technical solutions

Capabilities

QR code access control
Time tracking for logistics operations
Continuous use in a real production environment

Technical base

Backend APIsAsync workersObservabilityCloud infrastructure

Stack

Technologies and patterns I use most.

Grouped by layer to show depth without turning the section into a wall of logos.

01/backend

Backend

APIs, services, and integrations.

Node.jsTypeScriptExpress

02/frontend

Frontend

Web interface and delivery layer.

ReactNext.jsTailwind

03/database

Database

Persistence and consistent data access.

PostgreSQLPrismaSupabase

04/infra

Infra

Deploy, runtime, and environments.

AWSRenderVercel

05/observability

Observability

Logs, tracing, and monitoring.

logstracingmonitoring

06/architecture

Architecture

Queues, workers, events, and access control.

queuesworkerswebhooksRBAC

About

Backend, integrations, and production stability.

I do my best work on systems that need clean integrations, predictable operation, and straightforward diagnosis.

My focus is backend and integrations that cannot fail silently. I like organizing async flows, improving observability, and making systems easier to trust for both developers and operators.

Contact

Contact for roles, technical conversations, or product context.

The channels below are the fastest way to review code, reach out, and keep the conversation going.

If the role or technical context is a fit, this is the best place to start.

Channels

GitHub, LinkedIn, and the PDF resume are already available here.

GitHub

View GitHub

Repositories, experiments, and public technical history.

Open GitHub ->

View LinkedIn

Professional profile with background, experience, and career context.

Open LinkedIn ->

Send email

The fastest channel for opportunities, technical context, or product discussion.