Software engineer focused on backend product systems, SaaS operations, and applied AI with explicit boundaries.
PT-BR: aberto a oportunidades no Brasil e a trabalhos remotos com foco forte em backend.
- Product systems with approval paths, billing states, queue-backed workflows, and operator-facing docs.
- APIs that need contracts, verification steps, and reliable execution after launch.
- Applied AI features with budget limits, audit trails, and deterministic fallbacks.
QuoteFlow BR: a quoting engine for WhatsApp-first sales with versioned proposals, approvals, PDF output, and audit history.AcessoQR Public Core: operational tooling for QR-based access, offline validation, duplicate protection, and stronger field workflows.
| Repository | Why it matters |
|---|---|
| VOWGRID-API | Trust layer between AI agents and real-world actions, with simulation, policy evaluation, approvals, receipts, and rollback visibility. |
| OnboardPulse | Multi-tenant SaaS onboarding product with tenant boundaries, billing flows, follow-up jobs, storage modes, and AI budget controls. |
| MailSieve | Lean product API with OpenAPI contract, key flow, verification scripts, and deploy-ready operational discipline. |
| rede-neural-do-zero | Educational repo that still shows engineering rigor through deterministic evaluation, logs, tests, and reproducible outputs. |
- AI Guardrails: Simulation before execution
- SaaS Operations: Billing webhooks need replay discipline
- Backend Systems: Contracts beat clever APIs
- All notes: Writing index
- I prefer systems that stay explainable under real constraints: money, auth, retries, auditability, and deployment friction.
- I document trade-offs, failure modes, and setup paths because those details matter after the first build passes.
- I care more about clear boundaries and believable operations than about shipping impressive-looking demos.
- Portfolio: saviocodes.github.io/saviofilho.dev
- LinkedIn: linkedin.com/in/savio-filho-7a0212309
- Email: codessavio@gmail.com