Introduction to AI Automation with n8n & LangChain (no-code)

Build no-code ai automations with LangChain to transcribe audio from Google Drive, store text files, and trigger on new files to summarize, translate to Japanese, and generate audio.

Launch a no-code workflow with a manual trigger to fetch an MP3 from Google Drive, transcribe it with OpenAI, convert the transcript to a text file, and upload it back.

Build a no-code ai workflow that transcribes an mp3 to text, summarizes the content with a LangChain summarization chain, translates it into a language from 55 options, and generates audio.

Create a no-code n8n flow that watches a Google Drive folder for new MP3 files, downloads them, transcribes to text, and saves the results in a text transcription folder.

Learn to build a no-code chat with your documents that answers HR policies from PDFs and text files, with embeddings, vector stores, and database choices.

Set up your current vector store by creating a cluster on cloud.google.com, saving the cluster URL for credentials, and generating an API key token in data access control.

Demonstrates a no-code ai automation flow that retrieves an HR policy document from google drive, splits into chunks, creates embeddings with OpenAI, and stores vectors for querying.

Build a querying workflow using a chat trigger, vector store retriever, and embeddings to fetch similar HR policy documents and answer questions with a 3.5 turbo model.

Extend the document app with Slack integration to create a no-code HR policy advisor, using webhook, slash commands, and a Supabase vector store in a test workflow.

Create a no-code Slack integration workflow with a webhook and an HTTP request, linking a retrieval QA chain and vector database to deliver LLM-driven responses.

Introduce the traveler copilot, an ai-powered telegram bot for speech and image translation across 55 languages, with a node-based workflow for audio and image translation.

Set up a telegram bot with bot father, create traveler copilot bot, copy its token, then obtain an anthropic api key to access cloud three models with vision.

Create a telegram-triggered flow that downloads voice messages and transcribes them to text. Translate to Japanese, convert to speech, and send both audio and text back via Telegram.

Build an image translation workflow in a no-code ai automation setup using a telegram node to fetch and translate an image to the specified language with a basic lm chain.

Leverage a no-code AI workflow to analyze SEC 10-K data and generate a stock analysis report with a team of analysts, editors, and a publish step.

Walks through upserting a Nvidia SEC 10-K pdf into a vector store via a no-code template, using Google Drive and a webhook for Q&A agents.

Build a no-code ai automation workflow in n8n and LangChain with a settings-driven senior research analyst persona, integrating sec 10-k data via webhook and generating json research sections.

Explore delegation and split nodes that assign prompts to research analysts to produce report sections. Use a senior researcher plan and GPT 3.5 turbo to assemble coherent output.

Merge and senior editor synthesize research from multiple agents into one article, then paraphrase and polish for the investor community and publish as no-code markdown file to google drive.

Extend automated stock analysis by letting AI agents call tools via large language models. Use Financial Modeling Prep to fetch financial metrics, price targets, news, and earnings transcripts.

Build a custom financial metrics tool in a no-code n8n and LangChain workflow to fetch metrics via a ticker and API call, then include balance sheet and income statement analysis.

Add an earnings transcript tool to a no-code workflow using n8n and LangChain, call an earnings transcript API, and summarize analyst questions and management responses with a large language model.

Build an end-to-end AI vision powered PDF chat app that extracts complex PDF data, including tables and diagrams, into a high-quality vector store for chat-based insights.

Transform complex pdfs into concise blog posts via ai automation with a low-code workflow, using a four-agent team to extract content, plan sections, assign writers, and polish the final article.

Build a no-code ai automation pipeline that scrapes a website, extracts title, price, availability, image URL, and product URL, and loads data into Google Sheets using four nodes.

Explore building a technical analysis agent that analyzes Bitcoin charts using the relative strength index and directional movement index, generates chart outputs, and provides market insights with ai vision.

Harness an AI agent team—the research leader, project planner, research assistants, and editor—to produce a 1000-word, six-section, auto-citation article with citations using perplexity, then post it to ghost.

Transform spreadsheets into a knowledge base for natural-language search, comparative analysis, and q&a, using a no-code data agent to explore a 14,000-row stock dataset (debt, p e ratio) across years.

Create a visualization agent workflow that generates line, radar, and bubble charts comparing Google, Apple, and Microsoft on EPs, profit margins, and other financial metrics.

Learn to build a no-code ai workflow that links a telegram bot with a data analyst agent and a visualization agent, using deep seq v3 for cost-effective insights.

Automate 24/7 no-code news research with a team of AI agents monitoring Bitcoin, Nvidia, and SoFi to generate a citation-rich newsletter for Gmail.

Automate Reddit to newsletter curation with no-code ai research agents and auto citations, sourcing hot topics from the Side Project subreddit for newsletter generation.

Orchestrate a three-agent workflow—an orchestrator, a SQL data retrieval agent, and a visualization agent—to analyze and visualize energy data in Postgres, handle mislabeled data, and chart production growth 2023–2024.

Join the automation engineering school community to build real AI automation skills through course projects, participate in a soft launch as a founding member with a lifetime discounted rate.