Project Overview:
Designed and deployed "Odin," a full-stack, multi-agent AI operations assistant built on self-hosted n8n (Docker) and Python, paired with a custom React-based voice dashboard ("Odin Command Center").

The main dashboard of the ODIN Command Center, featuring the custom React-based "SPEAK TO ODIN" real-time voice interface connected to n8n secure webhooks.
Multi-Agent Architecture (5 Specialized Sub-Agents):
• Social Media Automation Agent: Automated content workflows, scheduling, and cross-platform social media management.
• WhatsApp Cloud API Agent: Automated customer inquiry handling, instant replies, and lead qualification.
• Airtable CRM Agent: Real-time lead capture, status tracking, and structured database updates without manual data entry.
• Gmail & Google Calendar Agents: Automated email triage, drafting, and smart appointment scheduling.

Backend multi-agent orchestration workflow built in self-hosted n8n, routing decisions through specialized sub-agents powered by Qwen Cloud Chat Model.
Voice & Dashboard Stack:
• Custom React Voice Interface: Real-time audio interaction connected to n8n via secure webhooks.
• Speech-to-Text & Text-to-Speech: Powered by Alibaba Cloud Qwen audio models (qwen-audio-3.1-asr-flash and qwen-audio-3.1-tts-next) for low-latency voice commands and audio responses.
• Self-Hosted Infrastructure: Deployed locally on Docker with modular JSON workflows, REST APIs, and robust error handling.

Operational dispatch view: Synchronizing workshop schedules with Google Calendar and sending automated WhatsApp updates directly from ODIN.
Result:
Unified social media management, WhatsApp communication, email, calendar booking, and Airtable CRM operations into a single voice-controlled AI command center.
0 likes · 0 views · Nov 16, 2026
Built "Odin," a Docker n8n & Python AI assistant with a React voice UI (Qwen STT/TTS) and 5 sub-agents for Social, WhatsApp, CRM, Gmail & Calendar.