Built by a developer,
for developers.

Hi, I'm Francisco.

I'm a software engineer from Cuba, based in Santiago, Chile, with 20 years of experience building production systems — Python backends, cloud infrastructure, financial platform integrations, and more recently, AI agent tooling.

Over the years I've worked at companies including Globant and Tata Consultancy Services, served as CTO at ServiSenior.com for four years, and built everything from distributed systems to MLOps pipelines on Google Cloud. I'm an AWS Certified Solutions Architect and hold Google Cloud certifications in infrastructure modernisation.

Why I built this

In late 2024 I started building AI agents seriously — not demos, but real autonomous workflows that take actions in the world. The same friction point kept appearing: agents need email addresses to register for services, receive webhook callbacks, validate flows in CI/CD, and interact with external systems.

Every existing temporary email service is designed for humans: disposable, ephemeral, no API, no webhooks, no programmatic control. If you're an agent or a test harness, you're on your own.

So I built UnCorreoTemporal — programmable temporary email infrastructure with a proper REST API, SMTP ingestion, Redis pub/sub, WebSocket streaming, and a full MCP Server so any LLM-powered agent can send and receive email like a first-class citizen.

What this is (and isn't)

UnCorreoTemporal is a developer tool, not a consumer product. It's for:

• Teams automating email-dependent workflows with AI agents
• Engineers testing email flows in CI/CD pipelines
• LLM agents that need isolated, ephemeral mailboxes to operate autonomously

Every technical decision — HTTPS/WSS enforcement, per-inbox isolation, short TTLs, API-first design — was made with autonomous agents and production reliability in mind.

Get in touch

Questions, feedback, integration help, or just want to talk shop about AI agents and infrastructure: