Embeddings: Mapping and Visual Search

Learn how embedding models automate product category mapping across marketplaces and power a visual search engine that detects subtle tile pattern differences.

Overview

This demonstration showcases the wide range of embedding models and illustrates their adaptability in different contexts through two main examples: first, automated category mapping that reduces manual effort by accurately matching products across different marketplaces; and second, an advanced visual search engine that detects subtle differences in images, such as nuances in tile patterns. These use cases illustrate how embeddings capture complex relationships and similarities, paving the way for innovative solutions in a wide range of domains.

Tech stack

Related projects

ProductBar: Using AI to identify ideas in customer feedback

Hamburg

A walkthrough of ProductBar, an AI system that categorizes customer feedback, demonstrates its analysis workflow, and explores implementation…

How we build the next generation embeddings and rerank model

Berlin

This talk explains the development of a state-of-the-art embeddings and rerank model that surpasses OpenAI's text-embedding-v3, enhancing AI…

Beyond the Prompt: Building Chatbots That Control Your UI

Hamburg

Learn how to integrate a chatbot that directly manipulates a frontend, using in‑chat UI components and automatic page…

Transforming large scale data into vector embeddings

Munich

Learn to convert large, complex visual datasets into domain-specific vector embeddings using a platform that speeds generation 20x…

Image-Aware Content Extraction - Building custom models for specific use cases

Berlin

This talk covers building custom layout detection models to extract and preserve relevant images from documents, enabling combined…

Blueprints, re-imagined

Cologne

Demonstrate a generative AI system that parses chemical plant flow diagrams, generates precise process descriptions, and includes background,…