If you want to learn:
- How to set up self-hosted n8n workflow automation on your computer using Docker?
- How to integrate OpenRouter with n8n to access advanced AI models like DeepSeek?
- How to create your first AI agent with chat capabilities and tool calling features?
- How to connect external APIs like MarketStack to your AI agent for real-time data access?
- What are the differences between using cloud-based AI models versus running Ollama locally?
- How to troubleshoot common issues with AI model integration and tool calling reliability?
Then this lecture is for you!
This lecture provides a comprehensive step-by-step guide to setting up a self-hosted n8n automation platform with AI-powered agents using OpenRouter and DeepSeek AI. You'll learn how to configure your n8n instance running in a Docker container, navigate the platform interface, and access key settings including version management and usage plans.
The tutorial walks you through creating your first workflow from scratch, starting with a chat message trigger and building an AI agent with memory capabilities. You'll integrate OpenRouter as your chat model provider using API key authentication, then configure DeepSeek v3.2 as your AI model to create an intelligent chatbot running entirely on your computer.
The lecture demonstrates how to enhance your AI agent by adding tools, specifically integrating the MarketStack API for real-time stock price data retrieval. You'll learn proper credential setup, parameter configuration, and how to enable filters for accurate data queries. The content covers troubleshooting tool calling issues with different models and explores alternatives like OpenAI's GPT o3s 120b model running through OpenRouter for more reliable performance.
You'll gain practical experience building AI-powered workflow automation that connects multiple apps and services, understanding the architecture of agents as LLMs equipped with tools to achieve specific goals. The lecture also discusses deployment options, including the differences between running Ollama locally versus using cloud-based solutions, and considerations for GPU acceleration on different operating systems.