Arlo multi-agent architecture illustration

AI / Systems Project

Arlo — Multi-Agent AI Assistant

CategoryAI / Systems Project
Tech StackPython, FastAPI, Telegram Bot API, OpenClaw, Claude, GLM-5
RoleSolo Developer
InterfaceTelegram Bot

Project description

A persistent AI assistant system built as a Telegram bot, powered by a multi-agent architecture. Routes tasks through specialized agents — each with a distinct role — for better quality and cost efficiency.

Uses PostgreSQL, runs on FastAPI. OpenClaw provides agent orchestration.

Architecture

Interpreter (Claude Haiku 4.5) — Fast routing layer.

Planner (Codex via OpenClaw ACP) — Architecture lead, breaks tasks into execution plans.

Executor (GLM-5 via OpenClaw) — Follows plans, writes code, runs commands.

Design principles

  • Cost-efficient routing
  • Deterministic execution plans
  • Persistent memory