Category: Automation & Integrations
-
AI Experiments in Supermarket Price Tag Recognition
Exploring Snowflake for Image Processing In January, I began experimenting with Snowflake Document AI to process supermarket price tag images. The goal was to extract structured product data from shelf photos, leveraging the scalable infrastructure and built-in machine learning capabilities of Snowflake. While the platform offered promising tools for document analysis, my early results showed…
-
Building an Automated Story Generation and Kindle Delivery System with Make.com
Sometimes you need to automate things which may be very helpful for your family. If they like to read stories and use Kindle – automation may be a gamechanger. Own stories delivered every day or week, directly to Kindle. Just spend some time (AI can assist) to create a list of stories and main idea/format about what you want…
-
Streamlining Document Processing with LLM Support: A Case Study
Application Overview The main goal of the application is to streamline document processing using LLM capabilities. Users can upload one or multiple PDF documents, specify categories for classification, and authorize requests using a unique code provided by the application owner. The LLM processes the documents based on commands like “categorize received document,” but this functionality…
-
Automating Content Creation and Publishing with Make.com and LLM
Creating and posting automated content to platforms like WordPress, Facebook and LinkedIn can significantly streamline your content strategy and save time. In this article, I’ll walk you through the process I used to create an automated solution for generating blog posts and sharing them on Facebook, using Make.com integrated with Perplexity PRO API and ChatGPT…
-
Exploring Snowflake DocumentAI for Supermarket Price Tag Recognition
Exploring Snowflake DocumentAI for supermarket price tag recognition, I experimented with extracting product names and prices from images. After setting up a schema and storage, I refined the input data by isolating price tags to improve accuracy. The model achieved 80% accuracy for product names but struggled with price recognition. Automating the process with queues…