Welcome to Vespa Voice, the podcast where AI leaders, search pioneers, and enterprise innovators converge. Each episode dives deep into the evolving landscape of AI, featuring candid conversations with experts shaping the future of agentic AI, search architecture, retrieval-augmented generation (RAG), and scalable enterprise applications. Whether you're a CTO driving digital transformation, a CIO reimagining data strategy, or an engineer building next-gen ML and search systems, this is your signal for what's next in intelligent infrastructure.
As AI gets more capable, its answers still depend on what it can find. Vespa CEO and cofounder Jon Bratseth explains why relevance goes beyond ranking to become a lasting competitive advantage. He also covers why agents need more than vector search, and how to deliver expert-level precision at the latency and throughput agents demand.
30 Oct 2025
How Hidden Search Problems Derail Great Search Products
This episode features Ravindra Harige, Founder at Searchplex. We explore a common but often misdiagnosed issue in modern software: search problems hiding in plain sight.
When users struggle with irrelevant results, poor UX, or high bounce rates, teams typically reach for the usual fixes—frontend tweaks, more testing, or scaling infrastructure. But in many cases, the real issue isn’t performance or design. It’s search.
We unpack why so many teams miss the mark, how conventional roles (frontend, backend, DevOps) obscure the need for search-specific expertise, and what it actually takes to build search that works.
10 Sept 2025
From Pilots to Precision: RAG and Search in Life Sciences
In this episode of Vespa Voice, host Bonnie Chase sits down with Harini Gopalakrishnan (former Global CTO of Life Sciences at Snowflake, now at Vespa) to break down how pharma and biotech are really using AI—from drug discovery and clinical trials to regulatory submissions and commercial launches.What you’ll learn:👉 From ML → GenAI → retrieval-first AI in life sciences👉 Why search is key to stopping hallucinations👉 Using multimodal data: text, images, proteins, genomics👉 How multi-phase ranking saves cost + boosts accuracy👉 Real-world pharma use cases: research, trials, commercialSubscribe for more deep dives into AI, search, and real-world innovation!
10 Sept 2025
The CXO Brief: Search Reimagined for AI Workloads
In this episode, Bonnie Chase sits down with Whit Walters, Field CTO at GigaOM, to dive into the newly released CXO Decision Brief on the future of search and AI infrastructure. They explore:- The risks of delaying modernization and how technical debt has become a competitive debt- Best practices for pilots, quick wins, and upskilling teams to adopt AI-native platforms- Where the market is heading in the next 2–3 years as companies move beyond Lucene-based engines- How Vespa’s architecture unifies real-time data with AI compute for next-gen use cases like Perplexity and Vinted
17 Jul 2025
Beyond Chat: The Hidden Use Cases Driving the GenAI Revolution
In this episode of Vespa Voice, Bonnie Chase sits down with Kamran Khan, CEO of Pureinsights, to explore the evolving role of search in the GenAI era. With decades of experience in enterprise search, Kamran shares how keyword search has evolved into hybrid, vector, and multimodal search—and how Retrieval-Augmented Generation (RAG) and agentic AI are reshaping enterprise applications.You’ll hear real-world use cases, including:- A “Shazam for cars” that uses audio search to diagnose engine issues- How publishers are enhancing subscriptions with semantic and generative search- Why search is foundational to GenAI—and what most enterprises get wrongKamran also offers hard-won advice on data readiness, the POC-to-production gap, and how to avoid chasing hype without a clear use case. Want more episodes like this?Subscribe to the Vespa Voice podcast and get new episodes delivered straight to your inbox: http://bit.ly/4kNRYqm
10 Jul 2025
From Minutes to Milliseconds: Vinted's Migration to Vespa Explained
In this episode, we sit down with Ernestas Poškus, Engineering Manager at Vinted, Europe’s largest platform for secondhand fashion. Ernestas shares how Vinted migrated from Elasticsearch to Vespa to overcome scalability challenges, reduce infrastructure costs, and improve both query and indexing latency—all while serving over a billion items.💡 Learn how Vinted:- Achieved 2.5x faster search performance- Cut infrastructure costs in half- Reduced indexing latency from 5 minutes to 5 seconds- Powers hybrid search and real-time recommendations- Runs dense retrieval, image search, and ML models at scaleWhether you're building large-scale search, exploring hybrid vector+text retrieval, or just curious about what powers Vinted’s smooth user experience—this is an episode you won’t want to miss.
24 Apr 2025
The Future of RAG: Search, Vision Language Models, and Agentic Workflows
In this episode of Vespa Voice, Bonnie Chase sits down with Jo Kristian Bergum to explore the paradigm shift in information retrieval—from keyword matching to hybrid search and now, to vision-language models (VLMs).Learn how VLMs and methods using ColPali are revolutionizing retrieval pipelines by enabling multimodal search across complex formats like PDFs, scanned documents, and visual-heavy manuals. From IKEA assembly guides to enterprise knowledge bases, Jo Kristian explains how "what you see is what you search" is unlocking new levels of context, accuracy, and efficiency.Watch to learn:- Why VLMs are essential for unlocking siloed data- How they outperform traditional OCR + text extraction workflows- What this means for RAG systems and agentic AI applications- Why CTOs can't afford to ignore this rapid shift💡 Don't miss this deep dive into the next frontier of AI-powered search.
22 Apr 2025
The Rise of Vector Databases: Inside GigaOm's Sonar Report
In this episode, we chat with Whit Walters, Field CTO at GigaOm, and dive into GigaOm’s latest Sonar Report on vector databases, exploring why Vespa.ai is recognized as both a Leader and Fast Mover in the space. We break down what sets Vespa apart—from its integrated architecture that runs data, indices, metadata, and machine learning inferences on the same physical nodes, to its unmatched performance at scale.
We explore how Vespa’s support for lexical, vector, and hybrid search methods gives enterprises strategic flexibility in tackling a range of retrieval challenges. With its ability to process hundreds of thousands of queries per second with low latency, Vespa is pushing the boundaries of what’s possible in enterprise-scale AI search.
Beyond vectors, we discuss how Vespa’s full-text search, structured filtering, ML-powered ranking, and real-time indexing combine to create a powerful, production-ready stack for next-gen search applications.
Contact Vespa Voice
Guest appearances
Books guests
Based on episode analysis; this does not confirm that the show is currently accepting guests.
Host of Vespa Voice?
Claim your podcast to manage its listing and keep your show details accurate.
Pod Engine is an independent podcast discovery and analytics service and is not affiliated with or endorsed by this podcast. Artwork and show content belong to their owners. Full legal notice.
Explore this show Podcast research with Pod Engine
