Hi All, We’re looking for support to build a prototype solution that can interpret photographed delivery sheets and map the extracted data into a structured Excel format. The project involves exploring both traditional OCR-based pipelines (e.g., Azure Form Recognizer, AWS Textract) and newer multimodal GenAI models (like GPT-4 with vision capabilities) to accurately capture fields such as delivery numbers, dates, quantities, and addresses.

Everyone’s talking about AI agents. But the truth? Most demos you see online are just that—demos. Even big players like Apple and Amazon struggle to make their AI features work in the real world due to issues like hallucinations and unreliable outputs. In this week’s video, I break down the differences between simple workflows and true AI agents and share practical strategies for building reliable AI systems, including: - How to use workflow patterns like prompt chaining and routing to solve real problems effectively - Why agent frameworks might not be the solution you think they are - The #1 thing you need to scale AI systems successfully (hint: it’s not a new tool) Learn how to move beyond the hype and build AI systems that actually work.

I did something pretty crazy in my last YouTube video. For a recent project that I worked on, I had to develop an AI WhatsApp bot. However, we needed to have complete control over the code and data, which meant we could not use any external services or platforms other than the WhatsApp API. So, I figured out how to build a WhatsApp bot with pure Python—nothing else! And I am giving it away completely for free, along with a video tutorial and all the code. (I have permission to share this, as I replaced the custom data part with a demo) You can easily sell this project for $5k+ to clients. Do you want to learn how to build and sell chatbots? Then check out the video below!

The goal of this group is to help you navigate the complex and rapidly evolving world of data science and artificial intelligence. This is your hub to stay up-to-date on the latest trends, learn specialized skills to turn raw data into valuable insights, connect with a community of like-minded individuals, and ultimately, become a Data Alchemist. Together, let's decode the language of data and shape a future where knowledge and community illuminate our way. Rules - Don't sell anything here or use Data Alchemy as any kind of funnel - We delete low effort community posts, and posts with poor English. Proofread your post first. - Help us make the posts high quality. If you see a low quality post, then click on the 3 dots on the post and "Report To Admins". Start by checking out these links - Classroom - Introduction - Roadmap - Contribution Be Aware of Scammers - Please be aware that this is a public group. Unfortunately, some people abuse the Skool platform to send DMs or post comments to trick people. This is the internet, so always do your own due diligence. Never automatically trust someone here on the Skool platform other than @Dave Ebbelaar's official account. To kick things off, please comment below, introducing yourself. Let us know: 1. Your name and where you're from 2. What project(s) you're currently focused on See you in the comments!

Recently went through this very addictive game Gandalf | Lakera – Test your prompting skills to make Gandalf reveal secret information. But this is more than just a game. Though the Gandalf challenge is light-hearted fun, it models a real problem that large language model applications face: Prompt Injection. Curious to know, have anyone tried this yet? I am stuck at level 7 now. I recommend to try this out. The game is pretty simple: AI has been told a password in its prompt 🔐- It has also been prompted to not reveal the password 🙊- You have to talk to AI and trick it into telling you the password 🤫 What are your views on this?