I wanted to give you a heads-up on an exciting new tutorial I’ve got in the works. This video is all about building a full-stack SaaS application that pulls in YouTube comments, processes them with CrewAI enterprise, and generates actionable content ideas. If you’re interested in building full-stack AI applications, this is for you! 🛠️ Here’s What You’ll Learn In this tutorial, I’ll guide you through each part of the process of creating this full-stack application. Here’s a breakdown: 1️⃣ Building the Frontend and Backend: We’ll use Next.js to set up the app and deploy it on Vercel. Plus, you’ll get hands-on experience connecting to a Neon Postgres database to manage data. 2️⃣ Integrating with CrewAI Enterprise: Learn how to harness CrewAI’s enterprise features to analyze YouTube comments, filter out casual messages, and focus on meaningful feedback. You’ll see how to create a system that automates data analysis and transforms raw comments into structured, actionable insights. 3️⃣ Generating Video Titles and Descriptions: We’ll configure CrewAI to turn filtered comments into potential video titles and descriptions. You’ll build a workflow that streamlines idea generation and content planning, using CrewAI’s advanced capabilities. 💡 Why This Tutorial is Worth Your Time This tutorial doesn’t just cover building one app—it teaches you how to apply the synthesize pattern for data processing, a core skill in AI development. Here’s why this matters: ✅ Real-World Adaptability: The synthesize pattern goes beyond YouTube. After learning it, you can apply it to dozens of other applications where large datasets need to be turned into insights. Imagine using this pattern for customer feedback, product analysis, or trend monitoring—there are endless opportunities to build your own AI-powered apps! ✅ Hands-On Full-Stack Skills: Get practical experience with tools like Next.js, Vercel, and Neon, and learn how to bring everything together into a seamless app. 📅 How to Catch the Release

Hey everyone! Welcome to the official support thread for all your questions related to the AI Marketing Platform course! If something isn’t clear or you’re stuck on a particular topic, feel free to ask away! Chances are, if you’ve got a question, someone else is wondering the same thing. I’ll be doing my best to respond as quickly as possible. This is also a fantastic space for you to connect with fellow course members—so if you know the answer to someone’s question, don’t hesitate to jump in and help out! We’re building a community where everyone can learn from each other, so let’s make the most of it! Looking forward to your questions and the discussions ahead!

@Brandon Hancock i agree, claude projects are the absolute shiznits. im almost 100% using claude vs openai when coding. gpt has taken me down too many rabbit holes or referenced too many deprecated/outdated products in its coding responses. claudes coding responses are not perfect but they are perfecter than gpt's. where i do find gpt to be outstanding is when i am looking for higher temperature responses. i use it almost exclusively when writing vsl's, cold email sequences, etc. i feel they that both products/companies understand where they excel and are leaning into that market differentiation

like the ugly step children we are, mac and windows os's are well represented but us linux nerds have to fend for ourselves. so, heres a good walkthrough of installing cursor on an ubuntu machine. im using v22 and it works fine except i cant get a desktop icon but thats a first world problem. btw, beware of fuse. it will wreck your day. toodles

just wanted to throw this out. in case anybody can use it... https://github.com/repoman1964/VAPI_LANGCHAIN_MIDDLEWARE_BASIC i am in love with vapi (for the most part). they've taken care of a lot of the heavy lifting when it comes to the telephony part of building agents. but i have had a lot of trouble getting llm's to behave when feeding queries through their integrations. talking to some people a lot smarter than me, the opinion is that my prompting is being mixed in with some vapi blackbox prompts and may be suffering from "lost in the middle". so i built this solution that gives you a lot more control over your llm's. its not low-code, mainly because low code tools dont have the ability to stream events (sse) back to the client (in this case vapi). at least i dont think they do. it would be cool but for now i think you actually need a real server. ive included both a flask and a quart version of the app in the repo. this first iteration is not chatable (no memory). i did that to keep things as basic as possible so you could see how we are langchaining back to vapi. and, because my langchaining skills are still iterating. i have a chatbot (conversation history) integration kinda working and the ultimate end state is to build a vapi agent with langchain. no explainer video yet but if enough interest id be happy to do walkthrough of the code. enjoy