Hey everyone! I’m thrilled to share an update on the course I’ve been creating for you. In case you missed it, I’m launching a series of courses focused on solving real-world problems using AI-powered apps. The ultimate goal? To help you become full-stack AI developers, ready to tackle the challenges of 2024 and beyond. **Progress Update: What’s New** In this first progress update, I’m excited to show off some of the new features you’ll learn to build in the course. These aren’t just any features—these are the kinds of tools that will give you a solid foundation as an AI developer. I’ve also implemented a few core AI and full-stack concepts that are absolute must-knows for anyone serious about AI development. **Demo & Next Steps** During the update, I also dive into a live demo where you can see these new features in action. Plus, I’ll cover some of the major enhancements I’m planning to build out next, so you know exactly what’s coming your way. I’m so pumped for you to see what we’ve been working on, and I can’t wait to hear what you think! If you have questions about anything in the video or the course as a whole, please drop a comment down below and I'll be happy to help! Cheers, Brandon 🧑‍💻🥂 P.S. If you missed the original course announcement and overview, you can watch it here: Click here to watch overview P.P.S. Here's the link to the course inside of the Skool community. Click here to view course

🇧🇷 👋 Hi Guys, for the past few weeks I'm trying to develop a rag model that can retrieve information from pdfs, but my first issue started with pdfs that contain charts and images. Does anyone knows how to extract all the data to a data frame using offline open source packages? And how can I retrieve this raw files together with the summarized embeddings?

So I was thinking of a script that looked at the task FIRST, and then designed and built the crew, and task and 'tools scripts' or just took the tools right out of langchain to do the task, without needing to set up a crew in python each time. Basically the script would use templates that would be populated with a crew of the LLM's choice, for given tasks, a list of tools made, and coded up by the CODING AGENT, whenever its built and the tasks all auto-populated. So the user basically types their idea, crew ai then has a set and structured buildcrew.py that is designed to build a crew to carry out the task... the structured 'build crew' would pretty much be all you'd ever need to build if it worked... (Big IF there right)

https://www.loom.com/share/2401d93407594e7091ac72dca6fcadac?sid=d7f6fd6a-23e6-4178-9f7e-4f90eb47131d https://www.loom.com/share/11222b1cc56a4bd7b8d5356d6f44a3ab?sid=5fd2c5d5-0a35-4cc2-ba34-d935970b78f1 Is that possible to mimic with crew, I want a bot that can go to google my business profiles of local businesses and then read like 20-30 reviews and identify the name of the business owner. Example: Many people leave reviews stating "john did a great job, will use his service again" So the bot will then identify JOHN as the owner. Can some one provide me some insights.!

Hi there, I'm trying to utilise multiple llm's to assign different tasks. Would be great if I could get some guidance on how to accomplish this.