• Overview of Vespa.ai

    Get a high-level introduction to Vespa.ai. In just two minutes, you’ll understand how Vespa is positioned as a platform built for performance, scalability, and accuracy—core requirements for powering modern AI-driven applications. Ideal for a quick orientation to what sets Vespa apart.

  • Vespa Architecture in 3 Minutes

    Learn how Vespa’s architecture delivers performance and accuracy at any scale. The video explains how vectors, tensors, text, and metadata are stored and updated in real-time, how computation is moved to the data to remove network bottlenecks, and how distributed processing keeps latency low even on massive datasets.

  • Getting Started with Vespa

    In this tutorial we walk you through the steps to build your first Vespa application in Python. The video is based on the getting started tutorial in the link bellow, so if you want more details, want to copy the code, or just prefer the notebook format then you can follow the link bellow!

    https://blog.vespa.ai/vespa-quickstart-how-to-set-up-an-application-with-vespa/

    The tutorial covers:

    • Setting up of your Vespa Cloud account
    • Preparing your data
    • Creating a Vespa application package
    • Configuring your schema
    • Deploying to Vespa Cloud
    • Feeding your data
    • Querying the application

    Link for creating Vespa Cloud account: https://console.vespa-cloud.com/

    Link to dataset: https://tinyurl.com/imdb100dataset

  • From Legacy to Leading Edge: Vinted’s Journey to Data Modernisation

    In this episode of the Don’t Panic, It’s Just Data podcast, Kevin Petrie, VP of Research at BARC and the podcast host, is joined by Dainius Jocas, Search Engineer at Vinted, and Radu Gheorghe, Software Engineer at Vespa.ai. They discuss how Vinted, an online marketplace for secondhand products, modernised its data architecture to address new AI search use cases and the challenges faced with Elasticsearch.

    From the switch to Vespa and the advantages of supporting multiple languages and complex queries, the podcast offers insights on the trade-offs organisations must think about when updating their search systems, especially regarding AI and machine learning applications.

     

  • Leveraging Data Beyond Text: Multi-Modal AI at Scale

    Text isn’t the only data worth searching. Learn how Vespa.ai powers multi-modal search by combining text, image, and video retrieval—all at enterprise scale. In this session, discover cutting-edge strategies for integrating embeddings, low-latency vector search, and adaptive filtering to build powerful search and discovery engines.

  • Vespa Demo: Real-Time AI for Smarter Product Discovery

    Watch how Vespa powers next-generation product search, personalization, and ranking in one unified system—built on real-time tensor-based retrieval. This short demo shows how semantic visual search, real-time user profile updates, and business-aware ranking come together to deliver fast, scalable AI experiences. From similar product recommendations to adaptive, preference-driven results, see how Vespa brings intelligent relevance to life.

  • Webinar: Unlock the Future of eCommerce – One Platform, Unlimited Possibilities

    In this webinar recording, learn how Vespa.ai transforms online retail by combining search, ranking, and recommendations in one scalable platform. See how Vespa’s architecture enabled Vinted to scale to over 1 billion listings, accelerate query speeds, and cut operational costs—all while delivering precise, AI-powered search experiences.

  • Webinar: Advanced Video Retrieval at Scale

    Text isn’t the only source of valuable information—images and videos are packed with untapped intelligence. But extracting that insight at scale, in real time, is no small task.

    In this webinar, we explore the latest breakthroughs in large-scale video retrieval, powered by Vespa.ai and TwelveLabs. You’ll discover how modern architectures are enabling high-performance, multi-vector video search—without tradeoffs in speed or scalability.

    What you’ll learn:

    • How Vespa and TwelveLabs use multi-vector retrieval to improve video search accuracy

    • Strategies for scaling video retrieval across massive datasets without sacrificing performance

    • Real-world applications across industries—from media to security to AI assistants

    Whether you’re building AI-powered content discovery or managing large multimedia archives, this session will give you a behind-the-scenes look at what it takes to make video searchable at scale.

  • Webinar: Scaling AI Beyond Vectors: Building Smarter, Faster Systems for RAG

    During this Data Science Connect webinar, Aakarsh Ramchandani, Chief Strategy Officer, Ravenpack (Bigdata.com) and Jon Bratseth, CEO and Founder of Vespa.ai discuss strategies for leveraging generative AI beyond vector databases. Topics include scaling AI systems for performance, combining machine learning with search capabilities for enhanced retrieval-augmented generation (RAG) and best practices for getting started with generative AI.

  • Webinar: Transforming the Future of RAG with ColPali

    During this Data Science Connect webinar, Aakarsh Ramchandani, Chief Strategy Officer, Ravenpack (Bigdata.com) and Jon Bratseth, CEO and Founder of Vespa.ai discuss how ColPali is transforming information retrieval by integrating visual data from multimodal documents like PDFs, unlocking new possibilities for Retrieval-Augmented Generation (RAG). This cutting-edge solution simplifies processing for visually rich documents, ensuring greater accuracy and relevance in AI-driven workflows.

  • TechTalk: Building a Visual RAG application with Vespa in Python

    This AI Camp meetup presentation describes how to build an end-to-end Visual RAG application for PDF files using Vespa, with a Python-only approach. Learn practical insights and best practices for developing your own Visual RAG applications, with a focus on scalability and performance. The session also covers promising advancements in document search and retrieval. All referenced code is open source, providing a valuable resource for participants to apply these techniques in their own projects.

  • Webinar: RAG and Roll: Transforming GenAI Use Cases

    Watch this on-demand webinar from industry analysts BARC, to explore how Retrieval-Augmented Generation (RAG) provides a simple, cost-effective alternative to fine-tuning GenAI language models for domain-specific data. Learn how RAG works, including its integration with language models and data pipelines, discover common architectural approaches such as vector, relational, and graph RAG, and gain practical insights and guiding principles to get started.