Recordings

Schedule

Join us in Berlin for Haystack Europe 2026 – an immersive experience featuring workshops and two days of conference talks tailored for search and AI professionals.

Monday, September 14

09:00
♥

Training 1: Building Modern Search Platforms for Humans & AI

Daniel Wrigley, René Kriegler
Training
09:00 - 13:00
Main Stage
Search is no longer just a product feature — it’s becoming a foundational platform that powers AI assistants, copilots, personalization engines, and internal knowledge systems. Shifting from a solid onsite search experience to a true search platform raises new questions: how should teams collaborate, what data needs to be exposed and in what form, and how do you design for both human users and AI agents at once?
Add to Calendar (.ics)
13:00
Lunch Break 13:00 – 13:45
Lunch break 13:00 – 13:45
13:45

Tuesday, September 15

08:30
Doors Open & Registration 08:30 – 09:00
Doors Open & Registration 08:30 – 09:00
09:00
09:15
10:00
10:45
Morning Coffee Break 10:45 – 11:00
11:00
♥

Hyperbolic embedding models for visual search

Matin Mahmood
Talk
11:00 - 11:45
Main Stage
We trained an open image-text embedding model in hyperbolic space, so visual search can use hierarchy as well as similarity. The talk shows how radius and aperture create hierarchy inside the embedding model, how that helps retrieve specific items from compositional queries, and where the approach fails.
Add to Calendar (.ics)
11:45
12:30
Lunch Break 12:30 – 13:30
Lunch 12:30 – 13:30
13:30
♥

Frankensteining LLMs

Fabian Klemm
Talk
13:30 - 14:15
Main Stage
TNG's R1T Chimera models reached over 10 billion daily tokens on OpenRouter. How can a small consultancy produce its own LLMs? In our case: by Frankensteining them. This talk motivates: you can adapt LLMs to your own needs without being a big research lab. Including theory, technicalities, and lessons learned.
Add to Calendar (.ics)
14:15
15:00
Afternoon Break 15:00 – 15:15
15:15
♥

Search (Re)platforming for Humans and AI

Daniel Wrigley, René Kriegler
Talk
15:15 - 16:00
Main Stage
Balancing business rules, vector search, and AI evaluation is a shared challenge. Modern search systems must now serve both human and agentic users. In this talk we share grounded insights on search replatforming, API-driven tools, deterministic query governance, and the growing complexities of evaluation in the age of generative AI.
Add to Calendar (.ics)
16:00
16:45
Short Break 16:45 – 16:50
Refreshment Break 16:45 – 16:50
16:50
16:55
♥

Your A/B test was a coin flip

Artem Lukanin
Lightning Talk
16:55 - 17:00
Main Stage
Lightning Talks at Haystack are short, five minute presentations to present a new idea, an interesting problem you’ve been working on, or showcase a novel approach. The program will be decided during the first conference day, so bring your ideas for lightning sessions to Haystack!
Add to Calendar (.ics)
17:00
♥

Dissecting ranker latency

Roman Grebennikov
Lightning Talk
17:00 - 17:05
Main Stage
Lightning Talks at Haystack are short, five minute presentations to present a new idea, an interesting problem you’ve been working on, or showcase a novel approach. The program will be decided during the first conference day, so bring your ideas for lightning sessions to Haystack!
Add to Calendar (.ics)
17:05
♥

LLM focused extensions for UBI

David Tippett
Lightning Talk
17:05 - 17:10
Main Stage
Lightning Talks at Haystack are short, five minute presentations to present a new idea, an interesting problem you’ve been working on, or showcase a novel approach. The program will be decided during the first conference day, so bring your ideas for lightning sessions to Haystack!
Add to Calendar (.ics)
17:10
♥

jina-embeddings-v5-omni

Florian Hönicke
Lightning Talk
17:10 - 17:15
Main Stage
Lightning Talks at Haystack are short, five minute presentations to present a new idea, an interesting problem you’ve been working on, or showcase a novel approach. The program will be decided during the first conference day, so bring your ideas for lightning sessions to Haystack!
Add to Calendar (.ics)
17:15
♥

Making LTR easier to work with in OpenSearch

Scott Stults
Lightning Talk
17:15 - 17:20
Main Stage
Lightning Talks at Haystack are short, five minute presentations to present a new idea, an interesting problem you’ve been working on, or showcase a novel approach. The program will be decided during the first conference day, so bring your ideas for lightning sessions to Haystack!
Add to Calendar (.ics)
17:20
♥

Qdrant Vector Search and Hybrig Routig

Andrei Cristea
Lightning Talk
17:20 - 17:25
Main Stage
Lightning Talks at Haystack are short, five minute presentations to present a new idea, an interesting problem you’ve been working on, or showcase a novel approach. The program will be decided during the first conference day, so bring your ideas for lightning sessions to Haystack!
Add to Calendar (.ics)
17:25
♥

Netflix Query Testing Framework

Ivan Provalov
Lightning Talk
17:25 - 17:30
Main Stage
Lightning Talks at Haystack are short, five minute presentations to present a new idea, an interesting problem you’ve been working on, or showcase a novel approach. The program will be decided during the first conference day, so bring your ideas for lightning sessions to Haystack!
Add to Calendar (.ics)
17:35
17:45
Get-Together 17:45 – 21:00
Get Together @ Vagabund 17:45 – 20:45

Wednesday, September 16

08:30
Doors Open & Registration 08:30 – 09:00
Doors Open 08:30 – 09:00
09:00
09:15
10:00
10:45
Morning Coffee Break 10:45 – 11:00
11:00
11:45
12:30
Lunch Break 12:30 – 13:30
Lunch 12:30 – 13:30
13:30
♥

The RAG Cost Curve: When an index beats live search

Simon Hearne
Talk
13:30 - 14:15
Main Stage
"Just let the agent search live" has a cost curve, so does a compressed index. The curves cross but most teams pick a side before doing the maths. We'll use open benchmarks on quantisation and refinement to prove what compression really costs you in recall, in answer quality, and in dollars against live search.
Add to Calendar (.ics)
14:15
♥

What Gets Retrieved Gets Reinforced: Gender Bias in Agents

Maria Lungu
Talk
14:15 - 14:35
Main Stage
If your AI agent treats Sarah and Michael differently for the same query, would you know? We built a two-layer detector, learned why retrieval counting drowns in catalog noise, and landed on counterfactual testing as the only reliable signal. You will leave with a concrete, practical approach to testing gender bias in your own agentic systems.
Add to Calendar (.ics)
14:35
♥

Who Gets to Decide? Designing Agency Back Into AI Search

Öykü Doğru
Talk
14:35 - 14:55
Main Stage
Drawing on behavioral UX, cognitive offloading and spatial thinking, Öykü will share practical design principles for making AI-powered search easier to understand, inspect and challenge: How can interfaces surface sources, uncertainty and alternative perspectives while helping people remain active participants in the decision-making process.
Add to Calendar (.ics)
15:00
Afternoon Break 15:00 – 15:15
15:15
♥

Introducing Per-Document Embedding Fine-Tuning

Philippe Bouzaglou
Talk
15:15 - 16:00
Main Stage
We introduce per-document fine-tuning, a method that updates individual embeddings with new information while keeping the embedding model frozen. For example, applied to product search, it enriches product image embeddings with abstract attributes such as material and use case, improving retrieval without costly and risky model fine-tuning.
Add to Calendar (.ics)
16:00
16:45