I decided to act on @Brandon Hancock 's advice and embark on my own YouTube journey 🚀 The channel focuses on how you can use AI to grow your business. I’ll be testing various APIs, automated workflows, web scraping techniques, and sharing real-life examples with you. So far, I’ve released videos covering topics like lead qualification, tracking buying signals, handling inbound inquiries, and more. This community has truly changed my career trajectory. Since joining in March, I’ve built 2 products, made my first internet dollars 🤑, and developed several AI agents for clients. I’ll still be active here, as I have so much to learn about development, AI, and more. Feel free to reach out if you want to chat 👋 Here’s the link to the channel: https://www.youtube.com/@vladoesgrowth/videos

I built an internal tool that helps me figure out how much to charge clients. I also use it as a lead-gen. FYI, I usually apply a 6 months payback rule. (ie. 50% of the money saved in the 1st year). Thought you might find it helpful. ps. tried to bring bit of fun with one of the case studies ;) https://estimate-ai.streamlit.app/

Hi everyone - What's to best practice to loop through tasks output to build off pydantic objects off of them? context: - I have an object ContentCreator that contains a object lists. (e.g. LifeEvents is a list of LifeEvent objects) - Each task is ragging a database to retrieve each object (e.g. different LifeEvent objects) - Each of these tasks returns a pydantic output (ie. ContentCreator) - But then when I try to call the output pydantic at the end of crew, I dont get the consolidated ContentCreator object. How come? What's the best way to do that?

How do you ensure that a task output does not regurgitate the example you set in the expected_output? I'm giving an example of a pydantic object in the expected_output of a task, and whenever the task doesn't manage to generate its own object, it tends to make up data, reusing the info from the example I provided in the expected_output. No matter what I say in the tool, task, and agent. Is there a way to ensure that the agents can more easily distinguish example outputs from expected outputs?

No matter what I do, my RAG agent will always get a reply from the ChromaDB. Does the App() instance can also hallucinate?