Vespa.ai Live

9-10 September 2026 at 📍 Lumiere London

An in-person event for the practitioners shaping the next generation of AI-powered search and retrieval systems.
This year's agenda includes speakers from Etsy, Walmart, Ravenpack, and more.

Get Your TicketAgenda
SpeakersWhere to Stay

Let the countdown to Vespa.ai Live begin

Get your Ticket

Agenda

Wednesday, September 9 (Pre-Event)

Vespa.ai Live Training Sessions

Join us the day before for hands-on Vespa training sessions:

Vespa 101: Perfect for beginners getting started
Ranking 202: A deeper dive into improving retrieval quality

Learn more and register

Thursday, September 10

8:00am: Badge Pickup and Breakfast

Pick up your badge, grab a hot or cold drink, and get conversations started early.

9:00am: Welcome

Join us as we kick off Vespa.ai Live.

9:10am: Keynote by Jon Bratseth, CEO, Vespa.ai

All You Need to Solve Everything is to Have the Right Information at the Right Time

Drawing from decades of experience building large-scale search and recommendation platforms, Jon will explore the architectural shifts driving today’s most advanced AI experiences across commerce, search, recommendation, and agentic applications.

Full Session Description

9:45-10:15am: Path to Success: How to overcome personal bias, engineering concerns and other pitfalls to take Vespa to production, Cem Ünsal (Walmart)

Cem talks about his personal search journey with Vespa starting in early 2000s. He will talk about his approach to making Vespa available for various verticals, emphasizing potential pitfalls and short-term gains. Names of the innocents will be redacted.

Read Full Session Description

10:15-10:45am: Vespa at Etsy, Gregg Donovan (Etsy, Inc.)

Etsy's Vespa adoption from agentic auto-research to PyVespa automation. A field report from our first year adopting Vespa for search and ads at Etsy.

Read Full Session Description

10:45-11:00am: Coffee Break
11-11:30am: Nuances of Binarized Embeddings-Based Retrieval (Dainius Jocas, RavenPack)

As vector databases move toward billion-scale datasets, the memory and computational overhead of the HNSW index becomes a significant bottleneck. This talk chronicles an engineering journey into the internals of Vespa, exploring a provocative question: Can we eliminate the HNSW index entirely for high-volume financial data?

Read Full Session Description

11:30am-12:00pm: PROBLEM-1 was Titled "help": Stories from the Trenches (Andreas Eriksen & Marlon Saglia, Vespa.ai)

At Vespa we keep a JIRA project called PROBLEM. Several hundred tickets in, we have a small archive of stories that taught us something — usually about ourselves. This talk is four of them.

Read Full Session Description

12-12:30pm: Trey Grainger (Searchkernel): Talk title will be announced soon.

More details to come.

12:30-1:30pm: Lunch

Enjoy a delicious lunch between sessions.

1:30-2:00pm: From Legacy Search to Vespa: What a Real PoC Taught Us (Christiane Lemke & André Charton, Kleinanzeigen/Adevinta)

Learn how Kleinanzeigen evolved from a legacy search-first relevance system to a Vespa-powered recommendation platform. This session shares real-world lessons from a production PoC, covering incremental migration, embeddings, attribute extraction, multimodal signals, and the challenges of modernizing retrieval and ranking with imperfect data.

Read Full Session Description

2:00-2:30pm: Autoresearch for Ranking: Honest Lessons (Doug Turnbull, SoftwareDoug LLC)

What I've learned from nearly a year of autoresearching search relevance.

Read Full Session Description

2:30-2:45pm: Coffee Break

Take a break before the next set of sessions.

2:45-3:15pm: Evaluating Search in the LLM Era: Tools for Data & Vectors (Alessandro Benedetti & Daniele Antuzi, Sease)

This talk explores two open-source tools that leverage large language models to help with search quality evaluation:
the Dataset Generator to generate queries and ratings from your corpus of information and the Vector Search Doctor to diagnose possible causes of the poor performances of your vector search implementation.

Read Full Session Description

3:15-4:00pm: Panel: Building Search for the Modern Age. Moderated by Charlie Hull

More details to come.

4:00-5:15pm: Barcamp

A series of short talks submitted on the day and selected by attendee vote.

5:15-5:30pm: Closing Summary

Join us as we wrap up Vespa.ai Live.

5:30-7:00pm: Drinks Reception with Snacks

Drinks, snacks, and networking with peers.

Meet Our Speakers

Jon Bratseth

Jon Bratseth, Founder & CEO, Vespa.ai

Jon is CEO and cofounder of Vespa.ai, and the architect and one of the main contributors to Vespa, the leading AI search platform. 

Speaker Details
Charlie Hull, The Search Juggler

Charlie Hull, The Search Juggler

Charlie is a veteran search and AI consultant with more than 25 years of experience helping organizations build scalable, high-quality search and retrieval systems. 
3-May-07-2026-06-09-37-3047-PM

Doug Turnbull, Retrieval Consultant, SoftwareDoug LLC

Doug is a Retrieval Consultant at Software Doug, LLC. He also authored Relevant Search and AI Powered Search.

Speaker Details
7-3

Gregg Donovan, Senior Staff Engineer, Etsy

Gregg is a Senior Staff Engineer at Etsy, where he works on search and machine learning.

Speaker Details
1-2

Dainius Jocas, Senior ML Architect, Ravenpack

Dainius is a Senior ML Architect at Ravenpack, mostly focusing on the scale and performance side of search.

Speaker Details
2-2

Trey Grainger, Founder, Searchkernel

Trey Grainger is lead author of the book AI-Powered Search and founder of Searchkernel. 
Christiane Lemke_1

Dr Christiane Lemke, Senior Engineering Manager, Adevinta/Kleinanzeigen

Dr. Christiane Lemke is a Senior Engineering Manager at Kleinanzeigen, where she leads engineering teams building scalable search and discovery solutions. 
Speaker Details
Andrè Charton

André Charton, Head of Search, Adevinta/Kleinanzeigen

André Charton is Head of Search at Kleinanzeigen, where he leads the development of large-scale search and discovery systems for one of Germany's largest online marketplaces. 
Speaker Details
8-3

Daniele Antuzi, Software Engineer, Sease

Daniele is a software engineer at Sease, passionate about high-performance data structures and algorithms.

Speaker Details
9-2

Alessandro Benedetti, R&D Director, Sease

Alessandro Benedetti is an Apache Lucene/Solr committer and Solr chair of the PMC, Director at Sease Ltd.

Speaker Details
4-4

Cem Ünsal, Distinguished Engineer, Walmart

Dr. Ünsal currently serves as a Distinguished Engineer at Walmart, where  he focuses on the strategic development of agentic AI ecosystems and scalable backend architectures. 

Speaker Details
6-3

Andreas Eriksen, Principal Software Engineer, Vespa.ai

Andreas has worked at Vespa for 15 years, doing product development, SRE/oncall, sales engineering, advent calendars, and whatever else needs doing.

Speaker Details
Speaker Headshots Vespa Live

Marlon Saglia, Principal Software Engineer, Vespa.ai

Marlon Saglia is a Senior Site Reliability Engineer at Vespa.ai. On a good day he keeps the platform up; on a worse day he writes the post-mortem.

Speaker Details
Modern office building representing our company headquarters

Where to Stay

Here are some suggestions for hotels near the venue. Prices range from £160-300 per night for dates in September. Of course, London has many other options for accommodation, but these are closest to the venue:

Can't make it to London?

You can still be part of Vespa.ai Live from wherever you are. Register for the livestream to watch technical sessions as they happen. You'll also have the opportunity to submit questions and have them answered live by speakers during the Q&A sessions.

Registration is required to access the livestream, so sign up today to reserve your spot and receive your viewing details before the event.

livestream
VespaAI-logo-green-RGB

Join us at Vespa.ai Live

Be part of the growing community driving the future of AI search, RAG, personalization, and recommendation systems.
Join us in person or reserve your spot to watch the livestream. 

Sign Up Now