Maique Moraes
Senior full stack engineer
based in Brazil
Maique Moraes · Senior Full Stack Software Engineer
I’m Maique Moraes, a full stack engineer with 6+ years building TypeScript products for enterprise procurement, logistics, financial services, and healthcare. I work end to end with React, Next.js, Node.js, and NestJS — and integrate LLMs, retrieval, and MCP tools into real business workflows.
→ ServicesBackends-for-frontend that compose data, check access, and fail clearly when essential data is missing.
Selected work
Professional work, described at a high level. The diagrams are original illustrations of the engineering ideas — not product screenshots or internal architecture.
Stratesys Enterprise procurement
A NestJS backend-for-frontend that turns supplier and proposal data from several services into one reliable view — and refuses to present an incomplete comparison as valid.
Grupo Inlog · Stratesys Applied AI in logistics and procurement
Two assistants, in logistics and in procurement, that answer from documents, read live operational data, and take actions — while the application keeps control of identity, permissions, and business rules.
Unicred Financial services · Frontend architecture
Migrating React features to single-spa microfrontends with clearer component boundaries, a shared design system, and tests that protect existing behavior.
Client names, internal systems, and proprietary details are intentionally left out.
Experience
From vehicle telemetry to enterprise procurement — with responsibility growing from individual features to backend integrations, architecture, quality, and helping other developers.
Current
Senior Full Stack Developer · Enterprise procurement and strategic sourcing for large companies · IntelligentBuy
Senior Full Stack Developer · Waste-collection management used by companies and public services in several Brazilian states · drivers, fleet, and routes
Senior Full Stack Developer · Digital applications for a nationwide financial cooperative
uFLY
Full Stack Developer · Healthcare allocation software built for one of Brazil’s largest hospitals
Built responsive interfaces for hospital workflows, integrated APIs, investigated production defects, and presented delivery progress to clients.
Autonomo Technologies
Full Stack Developer · Issuance and validation of official documents — driver’s licenses and international driving permits — used by state transit departments
Developed features with React, Vue.js, TypeScript, Node.js, and NestJS; refactored for maintainability, with unit tests, SonarQube, and code reviews.
Creare Sistemas
Backend Developer · from intern · Vehicle telemetry for multinational agribusiness companies
Where it started: investigated telemetry issues with Oracle SQL, JavaScript, and JSON, wrote training documentation, and trained new interns and a team of about ten non-developer colleagues.
Approach
Principles I’ve earned on real projects — each one tied to work described on this page.
When missing data could mislead a decision, an explicit, recoverable error beats a plausible-looking screen.
→ Bid comparison BFF
Retrieval answers from documents. Current status and actions come from the services that own them — with their permissions intact.
→ AI assistants
Clear boundaries between data, logic, and presentation give every change an obvious home — and an obvious test.
→ Financial platform migration
I used to over-design shared components for needs that hadn’t arrived. Now I timebox reversible decisions, ship the simple version, and save deep analysis for choices that are hard to undo.
→ A lesson from design-system work
Reviews, pairing, and training are how good decisions outlast the person who made them.
→ Mentoring and internal training
Field notes
Not every lesson comes from a big project. A few debugging stories that shaped how I work.
React · Performance
A vehicle list sent several identical requests on load and on every filter change. The effect depended on a filters object recreated on each render. I stabilized the dependencies so requests fire only when values actually change, and cancelled stale requests with AbortController so slow responses can’t overwrite newer searches.
TakeawayCheck reference equality before blaming the API — and rule out Strict Mode’s extra effect runs before calling it a bug.
Backend · Database
A routes-and-vehicles API slowed down as data grew. Backend logs showed one query for the list, then one per row for related data. I batched the related lookups, read the query plan with EXPLAIN ANALYZE, adjusted an index used by the filters, and verified that results, pagination, and ordering stayed identical.
TakeawayAn optimization needs an equivalence check, not just a faster number.
Delivery · Containers
Build passed, tests passed, image published — yet the new Azure Container Apps revision never became ready. Logs and revision events showed NestJS listening on port 3000 while ingress expected 8080. I aligned the configuration, verified health checks and endpoints, and reviewed other environments for the same drift.
TakeawayA successful pipeline proves the build, not the running service.
Read the full story: A green pipeline, an unhealthy appArchitecture · Collaboration
A teammate preferred running report exports inside the HTTP request; I was worried about large volumes. Instead of arguing, we tested different sizes. Larger exports were prone to timeouts, so they moved to background jobs with status tracking — and small exports stayed synchronous.
TakeawayA shared test settles a disagreement faster than a longer debate.
AI-assisted engineering
I work with AI agents every day to move faster — and I stay accountable for every line that ships.
I use Claude Code, Codex, and GitHub Copilot to explore codebases, investigate bugs, compare technical approaches, and speed up implementation, while keeping changes focused and maintainable.
AI helps me find edge cases, expand automated tests, and drive browser-level checks with Playwright MCP. Generated code and tests are run, reviewed, and validated against expected behavior.
AI-assisted pull request reviews in Azure DevOps sit alongside CI checks, static analysis, and human review. AI surfaces issues earlier; correctness, security, and maintainability stay my responsibility.
See it in practice: this website was built with AI agents under my direction.The code, tests, agent instructions, and commit history are public.(opens in a new tab)
Toolkit
Also: JavaScript, Vue.js, Oracle SQL, Git, Scrum.
Background
Maique is one of the most qualified professionals I have worked with. What stood out most: he learns fast, has strong technical knowledge, understands business rules, and is deeply committed — often putting in extra hours to solve problems that weren’t his responsibility.
Ricardo Oliveira
Managed Maique directly
Contact
I do my best work with teams that care about the details — clear interfaces, dependable services, and AI that fits real business rules. If that sounds like your team, email is the fastest way to reach me.
Based in Brazil (UTC−3), with flexible hours for overlap with teams in the Americas and Europe.
Most of my production code lives in private company repositories. This site’s source is public. (opens in a new tab)