--- title: Becoming AI Native · Vespa.ai description: How organizations are putting their proprietary data to work with AI image: https://content.vespa.ai/hubfs/RGP/vespa-ainative-dossier/assets/og.jpg --- [Skip the film, read the document](https://content.vespa.ai/becoming-ai-native#doc) [![Vespa.ai](https://content.vespa.ai/hubfs/RGP/vespa-ainative-dossier/assets/VespaAI-logo-dark-RGB.svg)](https://vespa.ai) 1. The same model 2. Your data 3. One question, two answers 4. Retrieval at work 5. Your data, put to work [Talk to sales](https://vespa.ai/contact-sales/?source=Resources%20-%20Becoming%20AI%20Native&t1_source=Marketing&utm_source=vespa-resources&utm_medium=resource-page&utm_campaign=becoming-ai-native&utm_content=bar) ![Six different buildings drawn on paper, a tower, a columned hall, a domed hall, a low office, a factory and a shop, each with the same dark chip on its front.](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/assets/p1.webp?width=1600&height=900&name=p1.webp) foundation models ![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-ravenpack-ink.png?width=560&height=71&name=logo-ravenpack-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-thomson-reuters-ink.png?width=540&height=182&name=logo-thomson-reuters-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-elicit-ink.png?width=343&height=95&name=logo-elicit-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-metal-ink.png?width=560&height=143&name=logo-metal-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-deviantart-ink.png?width=560&height=146&name=logo-deviantart-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-vinted-ink.png?width=560&height=187&name=logo-vinted-ink.png) ![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-perplexity-ink.png?width=81&height=20&name=logo-perplexity-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-civsy-ink.png?width=71&height=29&name=logo-civsy-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-yahoo-ink.png?width=74&height=21&name=logo-yahoo-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-spotify-ink.png?width=71&height=22&name=logo-spotify-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-kleinanzeigen-ink.png?width=72&height=52&name=logo-kleinanzeigen-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-etsy-ink.png?width=73&height=36&name=logo-etsy-ink.png) ![The same six buildings, each with its own kind of data drawn above it: stacked sheets, a row of bound books, loose pages with a chart, a locked file box, framed pictures and a rail of clothes.](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/assets/p2.webp?width=1600&height=900&name=p2.webp) billions of financial documents two million active U.S. statute documents scientific literature deal documents over 650 million works one billion searchable items ![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-ravenpack-ink.png?width=560&height=71&name=logo-ravenpack-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-thomson-reuters-ink.png?width=540&height=182&name=logo-thomson-reuters-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-elicit-ink.png?width=343&height=95&name=logo-elicit-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-metal-ink.png?width=560&height=143&name=logo-metal-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-deviantart-ink.png?width=560&height=146&name=logo-deviantart-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-vinted-ink.png?width=560&height=187&name=logo-vinted-ink.png) ![One question goes to two answer cards headed by the same chip, a grey card with a clock and an empty dashed tab under it, and a mint card with three source tabs under it, above a foundation and a spread of documents.](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/assets/p3.webp?width=1600&height=900&name=p3.webp) stale data or inaccuracies the right information, at the right time and in the right context ![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-ravenpack-ink.png?width=560&height=71&name=logo-ravenpack-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-thomson-reuters-ink.png?width=540&height=182&name=logo-thomson-reuters-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-elicit-ink.png?width=343&height=95&name=logo-elicit-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-metal-ink.png?width=560&height=143&name=logo-metal-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-deviantart-ink.png?width=560&height=146&name=logo-deviantart-ink.png)![](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-vinted-ink.png?width=560&height=187&name=logo-vinted-ink.png) ![Eleven mint tiles, each with its own icon and room for a name, stand on one long platform, and the tile with a shopping bag carries a small badge showing a rising bar chart.](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/assets/p4.webp?width=1600&height=900&name=p4.webp) Smarter investment research Better AI answers AI-powered research Meeting intelligence Better-informed public services Faster financial research More relevant marketplaces Personalized content discovery Smarter job matching More valuable advertising Evidence-based advocacy ![Twelve mint tiles stand on the one platform, the eleven from before, each with its own icon and room for a name, and a twelfth marked with a plus sign.](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/assets/p5.webp?width=1600&height=900&name=p5.webp) Smarter investment research Better AI answers AI-powered research Meeting intelligence Better-informed public services Faster financial research More relevant marketplaces Personalized content discovery Smarter job matching More valuable advertising Evidence-based advocacy eBook --- # Becoming AI Native How organizations are putting their proprietary data to work with AI [About this eBook123 wordsRead the section](https://content.vespa.ai/becoming-ai-native#d1) --- ## Data Is (Still) the New Oil Putting proprietary data to work means reliably retrieving and ranking the right information for the right user or application, at the moment it matters and at production scale. [Data Is (Still) the New Oil514 wordsRead the section](https://content.vespa.ai/becoming-ai-native#d2) --- ## “If you want to build a RAG-based tool, first build search.” — Adrian “Panda” Smith, Infrastructure Engineer, Elicit - [![Perplexity](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-perplexity-ink.png?width=57&height=14&name=logo-perplexity-ink.png)](https://content.vespa.ai/becoming-ai-native#story-perplexity) - [![RavenPack](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-ravenpack-ink.png?width=79&height=10&name=logo-ravenpack-ink.png)](https://content.vespa.ai/becoming-ai-native#story-ravenpack) - [![Thomson Reuters](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-thomson-reuters-ink.png?width=65&height=22&name=logo-thomson-reuters-ink.png)](https://content.vespa.ai/becoming-ai-native#story-thomson-reuters) - [![Elicit](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-elicit-ink.png?width=54&height=15&name=logo-elicit-ink.png)](https://content.vespa.ai/becoming-ai-native#story-elicit) - [![Metal](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-metal-ink.png?width=59&height=15&name=logo-metal-ink.png)](https://content.vespa.ai/becoming-ai-native#story-metal) - [![Civsy](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-civsy-ink.png?width=64&height=26&name=logo-civsy-ink.png)](https://content.vespa.ai/becoming-ai-native#story-civsy) [Customer stories, 1 to 6905 wordsRead the section](https://content.vespa.ai/becoming-ai-native#d3) --- ## Vinted — Better Product Discovery, Measurable Business Impact +1.1% transactions · +0.6% GMV · €3.5M+ additional merchandise sold - [![Yahoo](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-yahoo-ink.png?width=53&height=15&name=logo-yahoo-ink.png)](https://content.vespa.ai/becoming-ai-native#story-yahoo) - [![Spotify](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-spotify-ink.png?width=52&height=16&name=logo-spotify-ink.png)](https://content.vespa.ai/becoming-ai-native#story-spotify) - [![DeviantArt](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-deviantart-ink.png?width=58&height=15&name=logo-deviantart-ink.png)](https://content.vespa.ai/becoming-ai-native#story-deviantart) - [![Kleinanzeigen](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-kleinanzeigen-ink.png?width=53&height=38&name=logo-kleinanzeigen-ink.png)](https://content.vespa.ai/becoming-ai-native#story-kleinanzeigen) - [![Etsy](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-etsy-ink.png?width=65&height=32&name=logo-etsy-ink.png)](https://content.vespa.ai/becoming-ai-native#story-etsy) - [![Vinted](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-vinted-ink.png?width=63&height=21&name=logo-vinted-ink.png)](https://content.vespa.ai/becoming-ai-native#story-vinted) [Customer stories, 7 to 12815 wordsRead the section](https://content.vespa.ai/becoming-ai-native#d4) --- ## What could better search unlock for you? Explore how the Vespa AI Search Platform can help you build relevant AI experiences with retrieval, ranking and real-time data in one platform. [Talk to sales](https://vespa.ai/contact-sales/?source=Resources%20-%20Becoming%20AI%20Native&t1_source=Marketing&utm_source=vespa-resources&utm_medium=resource-page&utm_campaign=becoming-ai-native&utm_content=plate) [Summary99 wordsRead the section](https://content.vespa.ai/becoming-ai-native#d5) ## The full document About this eBook The same model --- ## About this eBook Close ### About this eBook Becoming AI native is about more than adding large language models to existing applications. As foundation models become increasingly capable and accessible, differentiation shifts toward proprietary information and how effectively organizations can retrieve, rank and use it. This eBook explores that shift through real-world examples of retrieval at work. Drawing on organizations using Vespa.ai across RAG, AI agents, search, recommendations, personalization and product discovery, it shows how retrieval is being applied to very different business problems, from improving commercial outcomes and customer experiences to supporting AI applications operating across billions of documents. Together, these examples illustrate an important point: retrieval isn't simply another component of the AI stack. It is how organizations put their information to work and ultimately profit. [NextData Is (Still) the New Oil](https://content.vespa.ai/becoming-ai-native#d2) Data Is (Still) the New Oil Your data --- ## Data Is (Still) the New Oil Close ### Data Is (Still) the New Oil Long before generative AI, organizations competed on proprietary information and how effectively they used it. As AI becomes embedded in core business processes, that advantage is taking new forms. So far, becoming AI native has focused largely on adding LLMs to applications. But as foundation models become increasingly capable and accessible, the model itself becomes less of a differentiator. What increasingly sets organizations apart is their information and how effectively they put it to work. The challenge is making that information available to AI applications. LLMs need the right information, at the right time and in the right context, without delays, stale data or inaccuracies. That makes retrieving and ranking proprietary information critical to the quality of the AI experience. Putting proprietary data to work means reliably retrieving and ranking the right information for the right user or application, at the moment it matters and at production scale. ### Retrieval at Work While RAG, AI search and increasingly AI agents are creating new retrieval requirements, the same underlying capabilities also power recommendations, personalization, product discovery, advertising and other AI applications. More than 100 companies run applications on Vespa Cloud, alongside a global open-source community numbering in the thousands. Together, they are applying retrieval to an extraordinarily diverse range of business problems. And these aren't necessarily separate worlds. Organizations can use Vespa.ai for the applications they need today and extend the same retrieval foundation as new AI experiences emerge. A retailer might start with Product Discovery, for example, then add conversational or agentic commerce over time. #### Generative AI and RAG Examples - **Smarter investment research** – agents retrieve and connect relevant information across complex institutional knowledge. - **Better AI answers** — finding and ranking trustworthy information for AI-generated responses. - **AI-powered research** — helping researchers find and synthesize relevant information across large information collections. - **Meeting intelligence** — making conversations, decisions and commitments searchable and available to AI applications. - **Better-informed public services** — retrieving relevant government knowledge for AI-powered services and applications. #### Search, Recommendation and Discovery Examples - **Faster financial research** — retrieving and ranking relevant intelligence across enormous datasets. - **More relevant marketplaces** — matching shoppers with products across rapidly changing catalogs. - **Personalized content discovery** — matching people with content, creators and communities based on their interests. - **Smarter job matching** — combining multiple signals to connect candidates with relevant opportunities. - **More valuable advertising** — understanding intent and ranking relevant ads within milliseconds. - **Evidence-based advocacy** — making verified information easier to discover and use. These use cases may look very different, but they share a common requirement: retrieving and ranking the right information, in the right context, at the moment it is needed. The customer examples that follow show how that plays out in practice — from RAG across billions of financial documents and retrieval driven by AI agents, to personalized discovery, measurable commercial impact and trustworthy, evidence-grounded AI. They also show that this isn't a capability that emerged with generative AI. The same retrieval foundation has been supporting large-scale AI applications for decades and is now evolving to meet a new generation of AI experiences. [BackAbout this eBook](https://content.vespa.ai/becoming-ai-native#d1)[NextCustomer stories, 1 to 6](https://content.vespa.ai/becoming-ai-native#d3) Customer stories, 1 to 6 One question, two answers --- ## Customer stories, 1 to 6 Close - [![Perplexity](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-perplexity.jpg?width=78&height=20&name=logo-perplexity.jpg)](https://content.vespa.ai/becoming-ai-native#story-perplexity) - [![RavenPack](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-ravenpack.png?width=110&height=14&name=logo-ravenpack.png)](https://content.vespa.ai/becoming-ai-native#story-ravenpack) - [![Thomson Reuters](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-thomson-reuters.png?width=69&height=24&name=logo-thomson-reuters.png)](https://content.vespa.ai/becoming-ai-native#story-thomson-reuters) - [![Elicit](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-elicit.png?width=74&height=21&name=logo-elicit.png)](https://content.vespa.ai/becoming-ai-native#story-elicit) - [![Metal](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-metal.png?width=74&height=21&name=logo-metal.png)](https://content.vespa.ai/becoming-ai-native#story-metal) - [![Civsy](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-civsy.png?width=63&height=26&name=logo-civsy.png)](https://content.vespa.ai/becoming-ai-native#story-civsy) ![Perplexity](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-perplexity.jpg?width=112&height=29&name=logo-perplexity.jpg) ### Perplexity — Building AI Search at Web Scale Perplexity is an AI-powered answer engine that combines web search with large language models to provide current, cited answers. Before it can generate an answer, it needs to find relevant information across a vast web collection, identify useful passages and assemble high-quality context for the LLM. Perplexity uses Vespa. Its broader search and answer experiences illustrate the demanding environment in which AI applications operate: enormous information collections, rapidly changing content and users who expect fast, relevant answers. As these applications evolve, retrieving the right evidence remains essential to the experience quality. Perplexity platform scale and performance: 780M+ queries/month · 200B+ URLs indexed · 200M+ queries/day · 358ms median Search API response time [Read the Perplexity case study](https://vespa.ai/case-studies/why-perplexity-chose-vespa-for-ai-search/) ![RavenPack](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-ravenpack.png?width=159&height=20&name=logo-ravenpack.png) ### RavenPack — AI Research Across Billions of Financial Documents RavenPack bigdata.com helps financial professionals turn vast amounts of information into actionable intelligence. Its Bigdata.com platform combines RAG with RavenPack's financial knowledge graph, allowing users to research and analyze billions of financial documents through a real-time AI research assistant. In financial markets, where information changes continuously, the challenge is not simply finding relevant content but retrieving it quickly and accurately enough to support timely decisions, while keeping answers traceable to their sources. RavenPack uses Vespa.ai as the search and retrieval platform behind Bigdata.com, supporting vector search across billions of documents. After several years using Vespa Open Source, RavenPack moved to Vespa Cloud to support its enterprise workflows and simplify the infrastructure required as the service scales. Vespa enables Bigdata.com to retrieve and rank relevant financial information at billion-document scale, providing the evidence its RAG applications need for real-time research and analysis. > “We transitioned to Vespa Cloud to support our enterprise workflows. This simplifies our infrastructure as we scale.” > > — Armando Gonzalez, CEO and Co-founder, RavenPack | Bigdata.com Billions of financial documents · Real-time AI research · OSS → Vespa Cloud [Read the Vespa/RavenPack announcement](https://vespa.ai/ravenpack-launches-bigdata-com-with-vespa-ai-to-revolutionize-billion-scale-vector-search-for-financial-research/) ![Thomson Reuters](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-thomson-reuters.png?width=96&height=33&name=logo-thomson-reuters.png) ### Thomson Reuters: Bringing Legal Search from Research to Production Legal search requires more than matching meaning. Results must reflect the right jurisdiction, legal authority and whether a ruling remains good law. For Thomson Reuters Labs, addressing this complexity also meant finding a simpler way to experiment, improve ranking and bring new search capabilities into production. Using Vespa, the team consolidated chunking, embedding, content enrichment and custom ranking within a single search application. A proof of concept covering more than two million active U.S. statute documents progressed into a production system serving customers, without rebuilding the research in a separate technology stack. **What Vespa enables:** Domain-specific retrieval and ranking, fewer separate services, and a more direct path from search experimentation to production. **2M+** active documents in initial corpus · Production deployment · Unified retrieval and ranking [Read the Thomson Reuters Labs blog](https://medium.com/tr-labs-ml-engineering-blog/building-legal-search-at-tr-4e4b597d595e) ![Elicit](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-elicit.png?width=106&height=30&name=logo-elicit.png) ### Elicit — Better RAG Starts with Better Search Elicit uses AI to help researchers find and work with scientific literature. Its engineering team argues that effective RAG requires more than putting embeddings into a vector database: if the underlying search can't reliably retrieve the right information, the language model has little chance of producing a good answer. Elicit uses Vespa to combine vector and keyword retrieval as part of that search foundation. Rather than treating vector search as an external memory for an LLM, Elicit approaches retrieval as a search problem — combining different retrieval techniques, structured queries and reranking to improve the quality of the information ultimately available to the AI application. > “If you want to build a RAG-based tool, first build search.” > > — Adrian “Panda” Smith, Infrastructure Engineer, Elicit [Read the Elicit blog](https://elicit.com/blog/search-vs-vector-db) ![Metal](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-metal.png?width=106&height=30&name=logo-metal.png) ### Metal — Retrieval Built for AI Agents Metal provides an institutional intelligence platform for private equity firms, connecting information across deal documents, expert calls, financial data, CRM records and other sources. Rather than treating this information as a collection of documents, Metal models companies, people, activities and financial data — and the relationships between them — so its AI applications can retrieve information in the context of an investment workflow. Metal uses Vespa.ai as the core retrieval layer for an increasingly agent-driven platform. AI agents determine what information they need, query different entity types, apply filters such as recency and business rules, and iteratively retrieve additional information before constructing the context required for an answer. Today, 95% of Metal's retrieval is performed by AI agents, with Vespa supporting the structured, multi-entity retrieval and ranking underneath them. > “95% of our retrieval is done by AI agents.” > > — Sergio Prada, Co-Founder & CTO, Metal [Read the Metal case study](https://vespa.ai/metal-ai/) ![Civsy](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-civsy.png?width=88&height=36&name=logo-civsy.png) ### Civsy — Making Evidence of Censorship Easier to Find Civsy is an AI-powered platform developed by Norwegian cultural rights organization Mimeta to document violations of artistic freedom around the world. Information about censorship was previously fragmented across local reports, media coverage and other sources, making incidents difficult to verify, compare and analyze. Civsy brings this information together into a searchable resource that helps researchers and advocates identify patterns that might otherwise remain hidden. Built on Vespa Cloud, Civsy combines structured information such as country, art form and restriction type with unstructured reports and other content. Hybrid retrieval helps users find both exact and conceptually related information across languages, while RAG generates summaries linked back to verified sources. As new reports are contributed, Vespa indexes and ranks the information in real time, helping Civsy maintain a current and transparent view of artistic censorship. From fragmented reports to searchable, verifiable evidence [Read the Civsy case study](https://vespa.ai/case-studies/mimeta-civsy/) [BackData Is (Still) the New Oil](https://content.vespa.ai/becoming-ai-native#d2)[NextCustomer stories, 7 to 12](https://content.vespa.ai/becoming-ai-native#d4) Customer stories, 7 to 12 Retrieval at work --- ## Customer stories, 7 to 12 Close - [![Yahoo](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-yahoo.png?width=74&height=22&name=logo-yahoo.png)](https://content.vespa.ai/becoming-ai-native#story-yahoo) - [![Spotify](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-spotify.jpg?width=72&height=23&name=logo-spotify.jpg)](https://content.vespa.ai/becoming-ai-native#story-spotify) - [![DeviantArt](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-deviantart.png?width=77&height=21&name=logo-deviantart.png)](https://content.vespa.ai/becoming-ai-native#story-deviantart) - [![Kleinanzeigen](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-kleinanzeigen.png?width=47&height=34&name=logo-kleinanzeigen.png)](https://content.vespa.ai/becoming-ai-native#story-kleinanzeigen) - [![Etsy](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-etsy.png?width=57&height=28&name=logo-etsy.png)](https://content.vespa.ai/becoming-ai-native#story-etsy) - [![Vinted](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-vinted.jpg?width=68&height=24&name=logo-vinted.jpg)](https://content.vespa.ai/becoming-ai-native#story-vinted) ![Yahoo](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-yahoo.png?width=104&height=31&name=logo-yahoo.png) ### Yahoo — AI Applications at Internet Scale Yahoo has used Vespa for more than two decades across search, recommendations, and personalization. What began as search technology evolved into a general-purpose platform for combining large-scale data, machine learning and real-time decisioning across Yahoo's consumer services. Historically, Yahoo reported around 150 Vespa-powered applications, serving approximately one billion users and processing around 800,000 queries per second. These figures illustrate the scale at which a shared retrieval and ranking foundation has supported many different AI-powered experiences, long before the emergence of generative AI. Proven deployment scale: ~150 applications · ~1 billion users · ~800,000 queries/sec [Read the TechCrunch article](https://techcrunch.com/2023/10/04/yahoo-spins-out-vespa-its-search-tech-into-an-independent-company/) ![Spotify](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-spotify.jpg?width=100&height=32&name=logo-spotify.jpg) ### Spotify — A Search Foundation for Content Discovery Helping listeners find relevant content requires more than matching keywords. People express their interests in different ways, and useful discovery depends on understanding both what they ask for and which content is relevant to them. Spotify adopted Vespa through an initial podcast search pilot and subsequently expanded its use beyond that application, including migrating workloads from Elasticsearch. The podcast example demonstrates how semantic retrieval can complement existing search methods, helping listeners find episodes even when their queries do not match the wording in titles and descriptions. Spotify’s broader adoption illustrates how an initial search application can become the starting point for a more widely used retrieval and ranking platform. [Read the Spotify Engineering blog](https://engineering.atspotify.com/2022/03/introducing-natural-language-search-for-podcast-episodes) ![DeviantArt](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-deviantart.png?width=109&height=29&name=logo-deviantart.png) ### DeviantArt — Personalized Discovery Across a Billion Documents DeviantArt, a Wix subsidiary, is one of the world's largest online art communities, with more than 100 million registered members and over 650 million works. With tens of thousands of new artworks uploaded every day and highly diverse user interests, discovery depends on continuously retrieving and ranking the right content while adapting to changing content, trends and user behavior. DeviantArt uses Vespa.ai to power personalized search and recommendations, combining machine learning, user context, custom query processing and real-time ranking within a single serving platform. The deployment manages more than one billion documents while delivering extremely low query latency, enabling DeviantArt to provide fast, personalized discovery experiences across its global community. > “Vespa gives us the flexibility to integrate custom ranking models, user context, and domain-specific processing within a single high-performance system.” > > Chris Nell, CTO, Deviantart 1B+ documents · Nearly 4,000 QPS · 12ms average query latency [Read the DeviantArt case study](https://vespa.ai/case-studies/wix-case-study/) ![Kleinanzeigen](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-kleinanzeigen.png?width=66&height=48&name=logo-kleinanzeigen.png) ### Kleinanzeigen — Reinventing Personalized Marketplace Discovery Kleinanzeigen is one of Germany's largest online marketplaces, where helping users discover relevant listings means understanding multiple and sometimes competing signals of intent. As part of rebuilding its homepage recommendations, Kleinanzeigen moved from Elasticsearch to Vespa.ai, bringing user profiles, retrieval and ranking together within the same serving platform. The new approach combines signals including click history, explicit preferences, recent searches and listing freshness to create more personalized recommendations. In Fashion, Kleinanzeigen found that simply combining these signals during ranking wasn't enough: relevant products first had to make it into the candidate set. By giving different signals their own retrieval paths and then combining and ranking the results, the team better reflected the diversity of individual user interests. A subsequent A/B test produced a “very substantial uplift in listing clicks” compared with the previous approach. > “You cannot rank what you did not retrieve.” > > — Kleinanzeigen Engineering [Read the Kleinanzeigen tech blog](https://medium.com/berlin-tech-blog/rebuilding-the-kleinanzeigen-homepage-feed-a-deep-dive-into-fashion-796ff379481a) ![Etsy](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-etsy.png?width=80&height=40&name=logo-etsy.png) ### Etsy: Unifying AI-powered product discovery Etsy consolidated five retrieval engines onto Vespa, simplifying a fragmented search architecture that had become difficult to maintain and improve. A shared platform brings retrieval and multi-phase ranking together, while partial updates make it easier to refresh embeddings and AI-generated content without rebuilding entire documents. Working with approximately **130 million documents**, Etsy’s teams can evaluate more candidates with sophisticated ranking models and share infrastructure, pipelines and evaluation tools. Applied scientists can focus on improving search, supported by automated research workflows that run unattended for weeks. The result is a simpler foundation for developing and testing AI-powered product discovery, with less engineering effort spent coordinating separate systems. [Watch the Vespa Live presentation](https://www.youtube.com/watch?v=INbNwOqoZFk) ![Vinted](https://content.vespa.ai/hs-fs/hubfs/RGP/vespa-ainative-dossier/img/logo-vinted.jpg?width=95&height=34&name=logo-vinted.jpg) ### Vinted — Better Product Discovery, Measurable Business Impact Vinted operates one of the world's largest second-hand marketplaces, with more than 120 million registered users and around one billion searchable items, handling up to 25,000 item searches per second. Search, recommendations and personalization help buyers discover relevant products across a huge, constantly changing inventory. As the marketplace grew, Vinted's Elasticsearch infrastructure became increasingly costly and difficult to scale, prompting the company to move its search and discovery workloads to Vespa. The move delivered improvements across performance, efficiency and relevance. Search latency improved by 2.5×, the infrastructure footprint fell from 120 to 60 servers, and Vinted increased ranking depth by more than threefold. More importantly, better search translated directly into business results: purchased transactions increased by 1.1% and Gross Merchandise Value by 0.6%, representing more than €3.5 million in additional merchandise sold through the platform. +1.1% transactions · +0.6% GMV · €3.5M+ additional merchandise sold [Read the Vinted case study](https://vespa.ai/vinted/) [BackCustomer stories, 1 to 6](https://content.vespa.ai/becoming-ai-native#d3)[NextSummary](https://content.vespa.ai/becoming-ai-native#d5) Summary Your data, put to work --- ## Summary Close ### Summary Across AI search, research, recommendations and product discovery, these stories show how organizations use Vespa to turn complex data into relevant experiences. The challenges vary, but common needs emerge: retrieve the right candidates, rank them for the task, and serve results as data and demand change. Vespa brings these capabilities together, helping teams simplify their systems, move ideas into production and improve the experiences they deliver. From Thomson Reuters’ legal search to Vinted’s measurable business gains, the examples show what better retrieval and ranking can enable. Tell us what you’re building and the challenges you need to solve. [BackCustomer stories, 7 to 12](https://content.vespa.ai/becoming-ai-native#d4)[Talk to sales](https://vespa.ai/contact-sales/?source=Resources%20-%20Becoming%20AI%20Native&t1_source=Marketing&utm_source=vespa-resources&utm_medium=resource-page&utm_campaign=becoming-ai-native&utm_content=drawer)[The reading editionBecoming AI Native](https://content.vespa.ai/becoming-ai-native-document?hsLang=en) ```json { "@context" : "https://schema.org", "@type" : "TechArticle", "about" : [ { "@type" : "Thing", "name" : "RAG" }, { "@type" : "Thing", "name" : "AI search" }, { "@type" : "Thing", "name" : "AI agents" }, { "@type" : "Thing", "name" : "Recommendations" } ], "author" : { "@type" : "Organization", "name" : "Vespa.ai", "url" : "https://vespa.ai" }, "datePublished" : "2026-10", "description" : "How organizations are putting their proprietary data to work with AI", "hasPart" : [ { "@type" : "WebPageElement", "name" : "About this eBook", "url" : "https://content.vespa.ai/becoming-ai-native#d1" }, { "@type" : "WebPageElement", "name" : "Data Is (Still) the New Oil", "url" : "https://content.vespa.ai/becoming-ai-native#d2" }, { "@type" : "WebPageElement", "name" : "Customer stories, 1 to 6", "url" : "https://content.vespa.ai/becoming-ai-native#d3" }, { "@type" : "WebPageElement", "name" : "Customer stories, 7 to 12", "url" : "https://content.vespa.ai/becoming-ai-native#d4" }, { "@type" : "WebPageElement", "name" : "Summary", "url" : "https://content.vespa.ai/becoming-ai-native#d5" } ], "headline" : "Becoming AI Native", "inLanguage" : "en", "isBasedOn" : { "@type" : "Book", "name" : "Becoming AI Native", "numberOfPages" : 11 }, "publisher" : { "@type" : "Organization", "name" : "Vespa.ai", "url" : "https://vespa.ai" }, "url" : "https://content.vespa.ai/becoming-ai-native", "wordCount" : 2575 } ```