👋 Say Hello

I’m Vitor Alves,
Fullstack Software Engineer
Based in Brasil, SP.

I design and build scalable web platforms, from complex front-end applications to backend integrations, focusing on maintainability, performance, and real business impact.

Over the last 10+ years, I’ve worked on large-scale systems in education, logistics, and integrations platforms, owning products end-to-end and shipping features used in production by thousands of users.

10+ Years in Software Engineering

Enterprise Production Systems

End-to-End Ownership

Remote Experience

Experience

Senior Frontend Software Engineer

Domo

2026 - Present

Accelerated the migration of a legacy VB-based vehicle inspection platform across multiple U.S. states to a modern React and TypeScript architecture by deeply integrating Cursor and autonomous AI agents into the development lifecycle.

Engineered a library of custom AI skills markdown-based architectural prompts, to standardize and automate the generation of complex UI components, state management, and Redux workflows across the team.

Increased development velocity using AI to rapidly implement complex multi-tenant frontend requirements, including automated form validation, dependent dropdowns, and responsive grid layouts within a shared monolithic repository.

Collaborated in a cross-functional Agile team, championing AI-assisted unit testing with Jest and React Testing Library to ensure high reliability while maintaining a rapid release cadence.

Senior Fullstack Software Engineer

Uber

2022 – 2025

Sole engineer responsible for the front-end integrations platform used to design and execute complex workflows using Apache Camel routes.

Built Medley, a visual platform to manage routes, from scratch, refactoring a legacy codebase into a clean, scalable architecture.

Creation of Overryde app, enabling JSON-driven application generation, including an OpenAI-powered routes generator and Udeal to help HR deal with merge and acquisitions.

Contributed across front-end and backend to improve workflow reliability and developer experience in Uworc.

Senior Fullstack Software Engineer

Luiza Labs

2019 – 2022

Worked on large-scale supply chain solutions, focusing on performance, reliability, and distributed systems.

Led the rewrite of a critical scheduling system from a Ruby on Rails monolith to a modern microservices architecture using Python (Flask), REST APIs, Google Pub/Sub, Kubernetes, and automated testing.

Built and maintained front-end applications with React, ensuring consistency, performance, and test coverage.

Fullstack Software Engineer

Eddydata

2015 – 2019

Developed and maintained enterprise web applications using Java EE and Spring Boot.

Worked on a large academic management system covering financial, student, and administrative modules.

Built a mobile application using Ionic to deliver notifications, grades, and academic information to students.

Projects

Koinonia: Devocional Bíblico

Flutter Dart Riverpod SQLite RevenueCat

Premium devotional app for daily Christian practice: guided meditations by theme, liturgy with audio, offline Bible study, a spiritual journal, and faith communities, all without ads or distracting feeds.

Designed an offline-first experience with parallel Bible translations, themed reading plans, home screen widgets, and share-ready verse cards for social media.

Shipped subscriptions via RevenueCat, an AI theological guide for deeper study, and a marketing site available on iOS and Android.

Website App Store Google Play

IdeaGen

Next.js FastAPI OpenAI Clerk Upstash Redis

AI-powered business idea generator for the agent economy. Authenticated users stream structured ideas covering problem, ICP, MVP, moat, risks, and go-to-market, with an optional novelty and feasibility score.

Built as a full stack app on Vercel with Next.js and FastAPI, Clerk auth and billing, OpenAI SSE streaming, and server-side request quotas with reserve-then-refund so credits are only spent on successful generations.

Live Demo GitHub Repo

Digital Twin

Next.js FastAPI AWS Bedrock Terraform

A live AI digital twin with persona-backed chat and session memory, deployed as a serverless AWS stack. The browser talks only to the API while Bedrock inference, conversation memory, and quotas stay on AWS behind least-privilege IAM.

Provisioned with Terraform and shipped through GitHub Actions using OIDC, with CloudFront, API Gateway, Lambda, S3 memory, and a DynamoDB-backed daily chat budget for predictable spend.

Live Demo GitHub Repo

MediNotes Pro

Next.js FastAPI AWS Lambda DynamoDB Clerk

Portfolio demo that turns consultation input into streamed AI clinical notes, then persists visits and exports Markdown or PDF. Built as a near-zero-cost AWS deployment with a containerized FastAPI app on Lambda, DynamoDB, S3, and Clerk auth.

Uses SSE streaming over a Lambda Function URL, lifetime rate limits to control OpenAI spend, and GitHub Actions with OIDC for ECR and Lambda deploys. Demo only, not for real PHI or clinical use.

Live Demo GitHub Repo

Decoville

Java EE Amazon EC2 Postgresql

Web application developed to manage inventory, customers, and rental contracts for companies that lease equipment and materials for events.

The system centralizes stock control, customer management, and rental workflows, ensuring accurate availability tracking and contract handling throughout the rental lifecycle.