Master AI Workflows: Zero to Hero with n8n & Cursor

This lecture will highlight the topics that are covered within Section 2, including signing up for n8n, a walkthrough of the user interface, as well as building three workflows together.

Students will sign up for n8n and participate in a walkthrough of the user interface. From the editor, official documentation, to the executions tab, and more, students will be familiar and comfortable with navigating the n8n user interface.

This lecture will introduce the three workflows that students will build in Section 2, including an Inbox Manager Workflow, a Personal Assistant Agent, as well as a Voice-Powered RAG system.

Students will create credentials and set up their Google Drive API Integration in n8n, as well as create a project within Google Cloud Console. By the end of this video, students will have the Google Drive API Integration enabled within n8n, and be familiar with how to enable integrations within Google Cloud Console.

In this lecture, students will follow a step-by-step video to build their very first automation workflow in n8n, the Inbox Manager Workflow.  From using Gmail to OpenAI, students will get their first experience working with API integrations. This workflow will be useful for learners of all levels and industries, whether for daily life or corporate inbox management, this workflow will provide value to all builders. The JSON file will be available for download in Resources as well.

Students will first watch a demo of the Personal Assistant Agent, followed by a build-along where every node will be reviewed in detail. From Gmail to Google Sheets, Google Calendar to Asana, this Personal Assistant Agent build is designed to help students take the next step in their building journey by implementing an AI Agent within their workflow. By the end of this lecture, students will have built a personal assistant that will be able to take actions in their inbox, calendar, as well as workspace.

This lecture will provide a beginner friendly introduction to Retrieval Augmented Generation, providing students with a foundational understanding of the final build concept within Section 2, the Voice-Powered RAG system.

This lecture will guide students in building an upserting workflow, positioning students to succeed with the final build in Section 2. Students will gain their first hands-on experience working with a vector database in Pinecone Vector Store, preparing them for the more advanced RAG techniques discussed in Section 3.

The final build in Section 2, students will learn to build a Voice-Powered RAG Agent. Working in unison with the Upserting Workflow from the previous lecture, students will learn to communicate with their data, using Telegram and Pinecone. By the end of this lecture, students will understand how to design upserting workflows, create RAG agents, and position themselves to best utilize the power of artificial intelligence.

This video outlines the key limitations of RAG and presents strategies to optimize its four core components—ingestion, query, retriever, and generator—with an emphasis on accuracy in business applications.

This video provides an overview of a previously built RAG system, detailing its components, ingestion flow using Google Drive and Pinecone, and query methods via n8n chat.

This video demonstrates how to configure the RAG system to use Slack in place of Telegram by replacing the corresponding nodes, enabling easier query handling across platforms.

This video walks through how to mention a custom Slack app and manage user messages, covering message storage and the use of AI nodes for automated responses.

This video explores the ingestion component of the RAG system, showcasing how Vectorize streamlines AI development with RAG-as-a-Service, intuitive ingestion tools, and a flexible, modular architecture.

This video covers the process of obtaining API tokens for Vectorize, Pinecone, and OpenAI, including login steps, token generation, and important setup tips like enabling billing on OpenAI.

This video guides you through setting up a full Vectorize pipeline, including file upload via source connector, Pinecone database integration, OpenAI model configuration, and saving index and namespace settings.

This video explains how to streamline the ingestion process by uploading files to Vectorize directly from n8n, allowing you to manage RAG agent uploads without manually accessing the Vectorize UI.

This video outlines an advanced, scalable RAG architecture, covering the ingestion pipeline and the use of metadata to enable hybrid search—semantic search through Pinecone and relational queries via Supabase.

This video demonstrates the advanced workflow where files from Google Drive are uploaded and processed, with resulting documents stored in Supabase and a vector database. It emphasizes verifying that metadata and content are correctly populated for reliable hybrid search.

This video explores how the RAG agent processes and responds to financial data queries, showcasing its ability to surface key metrics like revenue trends, profit summaries, and expense insights through natural language prompts.

This video offers a deep dive into the nodes of an advanced RAG workflow, showcasing how each node is configured and the role it plays in managing data flow, integrations, and overall system logic.

This video explains how to set up a feedback loop to automatically improve RAG agent prompts, featuring two workflows that evaluate and score outputs based on criteria like accuracy and relevance for continuous optimization.

This video provides an overview of the feedback loop workflow, offering context on how the system was set up to retrieve Loom video URLs based on user queries. It outlines what to expect in successful and unsuccessful runs, and explains how the LMS table and query logic work together to deliver relevant onboarding content.

This video showcases a live demonstration of the feedback loop system in action—pulling prompts from Airtable, generating feedback analysis, and storing the results to support ongoing iterations and continuous improvement.

This video provides a deep dive into the feedback loop system, breaking down each node’s configuration and role within the workflow—from pulling prompts and analyzing outputs to storing results—highlighting how the components work together to enable continuous refinement.

This video provides an overview of OAuth limitations, highlighting the challenges of managing API keys and access tokens in team environments. It introduces Nango as an open-source solution for simplified authentication and data sync, with a brief mention of Composio as an alternative for function calling.

This video provides an overview of the OAuth integration system, breaking down each section of the n8n workflow nodes used with Nango to manage authentication. It demonstrates how to connect accounts via a web app, securely store access tokens, and trigger workflows—offering an open-source starting point for implementation.

This video demonstrates the full OAuth integration system in action—extracting Loom video data from Airtable, passing it to Apify for transcription, installing the Slack app into a workspace for slash command interactions with your Loom videos, and enabling users to query specific videos. It also showcases a Google Drive integration to complete the end-to-end workflow.

This video walks through the setup process for configuring integrations within Nango, including how to define scopes, input your Client ID and Secret, and prepare your integration for use in OAuth flows.

This video demonstrates how to configure your Slack app via the Slack API by adding scopes, setting up webhooks in both Slack and n8n, and replacing traditional Slack nodes with webhook-based interactions. It highlights how this configuration allows you to deploy your Slack app into other workspaces through the web application flow.

This video provides a deep dive into the OAuth workflow in n8n, detailing the configuration and role of each node involved in managing authentication and token handling. It also covers the Slack RAG agent node setup, explaining how each node contributes to processing user queries and integrating with the broader system.

This video explores the evolution of coding tools through the lens of vibe coding, highlighting agentic IDEs like Cursor and Windsurf that enable high-level, conversational development with minimal supervision. It also introduces Model Context Protocols (MCPs) and how they enhance the functionality and flow of these IDEs, with a walkthrough on setting up your environment to get started.

This video serves as a step-by-step guide for setting up Model Context Protocols (MCPs) in the Cursor IDE, walking through how to structure your project, define context files, and optimize prompts to enable seamless interaction between your codebase and the AI assistant.

This video demonstrates how to build a functioning personal portfolio website using Next.js and Tailwind CSS, entirely within the Cursor IDE. It showcases how MCPs guide the AI to generate, structure, and refine the web application through high-level instructions and context-aware assistance.

This video showcases a Cursor IDE demo where a modern web template is imported and customized to match personal preferences. It walks through modifying design elements, layout, and content to transform the template into a unique, personalized website.

This video walks through how to create a new GitHub repository and push your code to it directly from the Cursor IDE, using the GitHub MCP to handle repository setup, commits, and version control through high-level prompts.