Hello World — Building ARIA in Public

1 min read EN

Part of: aria-progress

#aria #meta #ai

Hello World

Welcome to the Aethos Tech blog.

This is where we document the journey of building ARIA — an AI executive assistant that runs on our local infrastructure, coordinates tasks, monitors projects, and helps us ship faster.

What to expect

  • ARIA Progress — weekly updates on what got built, what broke, and what we learned
  • Tech Deep Dives — how we built specific features (MCP servers, Ollama routing, Hub integrations)
  • Micro SaaS Learnings — product and technical lessons from building in the Brazilian market
  • Essays — occasional thoughts on dev tools, AI, and indie hacking

The stack behind ARIA

ARIA is built as a Model Context Protocol (MCP) server that runs alongside Claude Code. It connects to:

  • A custom Hub (Next.js + PostgreSQL) for tasks, insights, and activity
  • Ollama for free local LLM calls (commits, translations, summaries)
  • Docker for infrastructure monitoring
  • WhatsApp for mobile-first interactions

More posts coming soon. Stay tuned.

// ARIA greets you every morning
const { greeting } = await aria_context();
console.log(greeting); // "Bom dia, César!"