IMPORTANT NOTE: Timing of sessions and room locations are subject to change.
The Sched app allows you to build your schedule but is not a substitute for your event registration. In order to attend OpenSearchCon Europe 2026, please visit our website to register.
This schedule is automatically displayed in Central European Summer Time (UTC+02:00). To see the schedule in your preferred timezone, please select from the drop-down located at the bottom of the menu to the right.
Sign up or log in to add sessions to your schedule and sync them to your phone or calendar.
Off-the-shelf embedding models underperform on domain-specific queries, and "just fine-tune it" fails for proprietary terminology. Why? Standard fine-tuning can't teach models vocabulary they've never seen. The embedding gap costs 40-60% of search relevance; while ANN indexing only costs 1-5%.
In this talk, we'll walk through end-to-end domain-adapted neural search—from embedding adaptation to index optimization. We'll examine a multi-stage pipeline (vocabulary extension → domain warmup → dictionary pre-training → multi-objective fine-tuning) with experiments showing each stage's impact.
But embedding adaptation is half the battle. When embeddings change, what happens to your index? We'll explore when changes require full rebuilds versus adaptive quantization—a mechanism using reconstruction error tracking to maintain optimal compression without costly rebuilds.
Attendees will learn why fine-tuning fails, diagnostic frameworks for embedding quality, and practical strategies for continuously-updated production indices.
Samuel is an OpenSearch Commiter, Maintainer, Technical Steering Committee member. He is currently the architect of DataStax OpenSearch offering and previously Architect Of OCI Search Services and OCI Observability.
Thursday April 16, 2026 11:10 - 11:50 CEST Bohemia 2
