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.
Join Carl Meadows, Chair of the OpenSearch Software Foundation Governing Board and OpenSearch Product Lead for Amazon Web Services, as he explores how OpenSearch is leading the transformation to agentic AI.
Discover how OpenSearch's vector database foundation and open-source community uniquely position the project for the agentic era. Carl will reveal how AI agents are democratizing search development—enabling developers to build production-ready applications in minutes without specialized expertise—and transforming observability for AI-powered systems.
With a growing base of contributors and accelerating momentum in 2026, the OpenSearch community continues to drive innovation. Carl will welcome new OpenSearch Software Foundation members, unveil new agentic tools, and share his vision for how OpenSearch is empowering builders to accelerate development, unlock new capabilities, and deliver the next generation of AI-powered search and observability solutions.
Carl Meadows is Director of Product at AWS for Amazon OpenSearch Service and is the board chair for the OpenSearch Software Foundation at the Linux Foundation.
Carl has over twenty-five years in product management developing enterprise software and cloud services. Prior to joining AWS, Carl held product management leadership roles at Engine Yard, SunGard Availability Services, SoftLayer/IBM and Symantec. He holds a BA from Berklee Colleg... Read More →
Thursday April 16, 2026 09:05 - 09:30 CEST Bohemia 2
Enterprise search is undergoing a fundamental shift. As user expectations rise and AI agents increasingly become primary consumers of search, the challenge is no longer simply finding information—but delivering the right context to drive better decisions, actions, and business outcomes. This keynote explores how enterprise search is evolving from keyword retrieval toward intent‑aware, adaptive systems that balance relevance, explainability, governance, and cost.
Drawing on IBM’s long‑standing commitment to open source and its growing leadership in the OpenSearch community, we present a practical framework for next‑generation enterprise search built around three forms of intent: query intent, journey intent, and product intent. We examine why “more AI” or vector‑only approaches are often insufficient, how traditional lexical techniques remain critical when blended with semantic search, and why understanding exploration versus exploitation is essential for designing effective search journeys and measuring ROI.
The session highlights how OpenSearch enables smarter context through hybrid retrieval, dynamic relevance, and observable, explainable pipelines—capabilities that are increasingly critical as search moves from human interfaces to agent‑driven experiences. We also explore emerging community trends, including agent builders and teams migrating from vector‑only systems, and preview upcoming OpenSearch innovations that support AI‑native, governed, and cost‑efficient enterprise search.
Attendees will leave with a clear mental model for modern enterprise search, actionable guidance for balancing innovation with fundamentals, and an understanding of why OpenSearch—developed in the open—plays a foundational role in the future of enterprise and agentic search.
Benchmarks produce clean numbers - but do they reflect what actually happens when users apply filters, facets, and hybrid queries in realistic document collections in production?
In this session, we evaluate the real-world performance of three vector search options available to OpenSearch users: Lucene’s built-in HNSW, the Faiss engine integration, and jVector, an open-source vector search plugin. While all engines publish compelling benchmark results and advertise strong scalability, most existing comparisons focus on isolated ANN performance rather than realistic query patterns.
Using identical datasets, queries, and filter selectivity, we examine how latency, recall, and throughput change once Boolean filters, aggregations, and hybrid retrieval are introduced. We break down where performance differences originate inside OpenSearch - including index construction behavior, memory usage, concurrency, and merge strategies.
Attendees will leave with practical guidance on when each engine is the right default choice, how to interpret vendor and community benchmarks critically, and how to design vector search evaluations that reflect real production systems.
Fernando is the CTO of Zeta Alpha, an Amsterdam-based startup helping High-Tech and R&D Enterprises take their Generative AI projects to production. With a PhD in theoretical physics and a hacker (builder) past, Fernando brings a unique blend of theoretical knowledge and hands-on... Read More →
Thursday April 16, 2026 10:20 - 11:00 CEST Bohemia 2
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
In this session, I'll cover the basics of building a diverse and inclusive community, either around new projects or creating safe spaces for projects that don't have one.
Inclusivity and equity and the keys to diversity, and science has shown time and again that diverse teams create better solutions. It's imperative especially now that we build community both in and out of software.
This talk will cover why community is important, how to jumpstart a community, and how to deal with growth and conflict.
Open Source Technology Evangelist, NetApp Instaclustr
Kassian (they/them) has been a coder for nearly two decades, and in devrel for over a decade. They bring an educational spin to DevRel and love to see that "ah-ha!" moment. They see every moment as a teaching opportunity, even if it's not very effective on their three cats.
Thursday April 16, 2026 12:00 - 12:20 CEST Bohemia 2
Before OpenSearch 3.1, plugins hand-rolled document-level access control, duplicating effort and creating inconsistent security behavior and user experiences. Even basic collaboration required security administrators to act as intermediaries.
To address this, OpenSearch introduced the Resource Sharing Framework, enabling secure, fine-grained, owner-driven sharing of plugin resources while complementing role-based access control. Since its introduction, the framework has evolved through OpenSearch 3.4 with more plugins onboarded, refined permission scopes, improved APIs, and operational hardening.
This session explains how the framework extends the Security plugin to protect high-level resources, allow owners (or full-access users) to grant and revoke per-user or per-role access at action-group granularity, and perform permission checks using standardized Java and REST APIs. You’ll see how the Anomaly Detection plugin adopts the framework to enable Google-Docs-style sharing without admin overhead. Attendees will leave with a practical migration blueprint for retiring bespoke access-control schemes and delivering a consistent, secure collaboration experience across plugins.
Craig has been engaged with the OpenSearch community since June 2022 and focuses on the Security of the platform. Craig has an extensive software engineering background of 10 years that includes managing the search infrastructure for e-commerce websites and healthcare institutions... Read More →
Experienced Software Developer with full-stack experience over 6 years. Currently working as Software Engineer on OpenSearch project as part of security plugin.
Thursday April 16, 2026 13:30 - 14:10 CEST Bohemia 2
OpenSearch users increasingly need flexibility in how and where their data is stored - querying data in cloud object storage without re-ingestion, sharing datasets across analytics tools and engines, or migrating between platforms without costly ETL pipelines.
This session presents a technical deep dive into how OpenSearch can now support indexing and querying vendor-agnostic open formats natively with the new multi-engine composable architecture.
The presentation covers how the composable storage layer abstracts format-specific details while maintaining query performance. We'll explore schema management across formats, predicate pushdown to minimize data scanning, and how still existing indices like Lucene can act as layer on top of open formats as secondary indices for accelerated filtering.
Through real-world scenarios and querying production data alongside cold-tier archives in object storage, sharing datasets with other analytics engines, and leveraging specialized formats like Lance for vector workloads or Parquet / Vortex for high-compression analytics, we demonstrate how open format support delivers data portability, interoperability, and freedom from lock-in.
Arpit Bandejiya is a Software Development Engineer ll at AWS, working in Opensearch. Over the past three years, he has focused on Shard management, scalability and performance optimization. His interests lie in distributed systems and building high performance scalable infrastruc... Read More →
Software engineer in Amazon , OpenSearch maintainer and leads features on improving the performance of indexing and search. Interested in designing high performance distributed systems.
This talk introduces Docling, an open-source document processing framework, and its integration with OpenSearch for production-ready AI applications. Docling converts complex documents such as PDFs, DOCX, and PPTX into structured, AI-ready formats while preserving layout, tables, images, and metadata for high-quality Retrieval-Augmented Generation (RAG). Docling is supported by the LF AI & Data Foundation. The session presents practical integration patterns for advanced search applications. We showcase a multimodal RAG architecture built with Docling and OpenSearch. By combining Docling’s document conversion with OpenSearch’s vector and metadata search, we enable visually grounded RAG systems where users can verify AI-generated answers in the source documents. Documents can be indexed by running Docling externally or by invoking Docling within an OpenSearch ingest pipeline using the Docling Java API. We also demonstrate an agentic application using OpenSearch agentic search and Docling MCP for document understanding and information extraction. Attendees will learn how these integrations simplify document ingestion and enable accurate and explainable AI applications.
Cesar Berrospi Ramis is a Senior Research Scientist at IBM Research, Zurich. With a background in mathematics and engineering, he specializes in AI applications in document understanding and is part of Docling's development team.
Aggregation performance in OpenSearch is often limited by JVM execution overhead rather than aggregation logic itself. Most queries process documents at a time (DAAT) through Lucene collectors and rely on deep virtual call chains, which restrict JIT optimizations and waste CPU resources at scale.
This talk explores how Lucene-aware, bulk-oriented execution can significantly speed up aggregation processing. We will show how batching document processing, simplifying collector and doc values access patterns, and reducing virtual method dispatch lead to more JVM-friendly execution paths. We will also touch on how Lucene skip indexes built on sorted doc values can be used to skip or bulk-process ranges of documents during aggregation, reducing unnecessary work when data can be summarized at a higher level.
Using real OpenSearch aggregation pipelines as examples, the session presents early performance results and discusses practical trade-offs when moving from fine-grained per-document logic to bulk processing. The talk is aimed at engineers working on search engines, Lucene-based systems, or JVM performance.
Ankit Jain is a Software Engineer on the Amazon OpenSearch Service team, leading performance and scalability initiatives for search infrastructure. He is an active maintainer and committer for the Apache Lucene and OpenSearch projects, with hands-on experience operating large-scale... Read More →
Are you interested in contributing to OpenSearch but unsure where to begin? This 30-minute presentation provides a practical roadmap for developers looking to make their first contribution to the OpenSearch GitHub project.
We'll start by exploring the OpenSearch repository structure and architecture, helping you understand how the codebase is organized across its core components. You'll learn how to set up your local development environment, including the essential tools, dependencies, and configurations needed to build and test OpenSearch successfully.
We'll also go into what expectations you should have from maintainers who review your code, as well as best practices for working with the community.
The IPCEI-CIS program is shaping the next wave of European cloud-edge innovation by enabling secure and sovereign digital infrastructures. In this context, Adeptic Reply demonstrates how natural language can be transformed into a fully structured deployment request. We built an open AI powered engine capable of structuring requests being able to be executed on federated cloud-edge sites, composed of a mix of direct-managed and third parties’ nodes. To keep these complex and inter-connected scenarios under control, we developed an observability layer allowing users to track performance deviations across the federated cloud-edge sites, ensuring reliability and transparency in the application lifecycle.
This session illustrates how structured AI workflows and federated observability come together to support the IPCEI-CIS vision and enable seamless, trustworthy deployment across distributed cloud domains.
Software Engineer working with modern technologies to build impactfulsystems. His experience includes challenging domains such as personal mobility enhancement, where technology can directly improve people’s independence and quality of life. He currently contributes to the 8ra initiative... Read More →
ICT Engineer graduated from Politecnico di Torino, with a solid background incomputer engineering and advanced expertise in AI and machine learning. Currently an AI Engineer at Adeptic Reply, contributing to Europe’s digitalsovereignty by developing cutting‑edge AI solutions within... Read More →
Friday April 17, 2026 09:15 - 09:30 CEST Bohemia 2
The Partner Roundtable is a great way to interact with OpenSearch leadership and have an open dialogue about the project. We will discuss strategy, challenges and opportunities, community and collaboration, future trends, training and enablement, and project growth. And of course, you are encouraged to openly exchange ideas and experiences, ask questions, share your thoughts, and engage in meaningful discussion.
By invitation only.
Friday April 17, 2026 09:55 - 11:05 CEST Bohemia 2
The current index authorization model in OpenSearch has a few quirks and weaknesses; these will be mostly noticeable when working in OpenSearch Dashboards or similar applications with users which have only limited access to indices. Thus, if you do not take care and forget to flip certain switches, you will encounter quite a few "Forbidden" errors in OpenSearch Dashboards. Over the past year, we have been working on a new index authorization concept for OpenSearch. The goal is a more logical and intuitive authorization model that provides a smoother and more predictable user experience.
This new approach is expected to be introduced as an optional feature in an upcoming OpenSearch release and is planned to become the default in OpenSearch 4.
In this talk, we will explain the new authorization concept, highlight its advantages, and provide practical guidance for cluster administrators on how to roll it out safely without causing service interruptions.
Freelance software architect with long track record in security and infrastructure software. Interested in finding optimal solutions which combine broad functionality with a user-centric view. Track record of many years in security software via collaborations with Search Guard and... Read More →
OpenSearch is welcoming but the path from first PR to maintainer can feel opaque. This session provides a practical, step-by-step playbook for contributing to OpenSearch and growing as a trusted maintainer.
Learn where to start (repos, good-first-issue), how to pick right-sized work, set up your environment, align with coding and testing norms, and craft high-signal PRs that are easy to review. We’ll cover collaboration habits—scoping changes, attending triages, avoiding churn, handling async reviews, documenting decisions—and show how to build credibility fast.
We’ll outline the maintainer path: observable signals (consistency, quality reviews, ownership, reliability, judgment), responsibilities (triage, design guidance, merges, releases, community health), and how nominations happen. Leave with a 30/60/90-day plan, a review-friendly PR template, and a roadmap to grow consistent contributions into visible impact.
Varun Bansal is a Software Engineer at Amazon OpenSearch Service. With 10 years of experience in software development and cloud technologies, he is passionate about developing and improving the OpenSearch Service, helping customers harness the power of open-source search and analytics... Read More →
This is the sequel talk to the earlier OpenSearchCon talk on the topic “Budget Friendly Semantic Search With Neural Sparse Search”. This follow up talk we will explain shortcomings such as 1/ pre-trained lack the domain knowledge 2/ off-the-shelf-domain-specific-sparse-model are either expensive/proprietary or non-existent
Level 300-400 talk covering, 1/ Intro to Neural Sparse Search 2/ Pre-trained model shortcomings 3/ Fine-tuning pre-trained model using open-source tools 4/ Use Search Relevancy Workbench to compare pre-trained and fine-tuned sparse model search results
A recommended pre-requisite is to watch the earlier presentation on “Budget Friendly Semantic Search With Neural Sparse Search” - https://www.youtube.com/watch?v=kx71KFf-Nv0
Of course its a completely “optional” pre-requisite as the previous talk will be condensed to a 5 minute lightning talk and will serve as the introduction
This talk will be demo-heavy, showcasing the fine-tuning possibilities of sparse encoder models using data from an existing OpenSearch index, which also means that the approach is suitable for even for those niche domains for which even proprietary models are not available in the market
Aswath Srinivasan is a Senior Search Engine Architect at Amazon Web Services currently based in Munich, Germany. With over 18 years of experience in various search technologies, Aswath currently focuses on OpenSearch. He is a search and open-source enthusiast and helps customers and... Read More →
Principal Specialist Solutions Architect, Amazon Web Services
Cédric Pelvet is a Principal Specialist Solutions Architect at AWS, focusing on AI and near-realtime distributed systems for data like OpenSearch, Kafka and Flink.
Friday April 17, 2026 13:35 - 14:15 CEST Bohemia 2
The Norwegian Labor and Welfare Administration (NAV) manages 1/3 of Norway’s state budget. Supporting their "product-first" paradigm for 20,000 employees required a shift from legacy monoliths to a centralized, self-service observability stack. NAV partnered with Aiven to migrate their logging infrastructure to a managed OpenSearch environment. Aiven has been a strategic partner of NAV by hosting some of NAV’s most critical data services for many years. As a founding member of the OpenSearch Foundation with active maintainers, Aiven provided a unique advantage in helping NAV adopt OpenSearch for their application logging. This technical deep-dive covers: * Migration Architecture: High-throughput logging from on-premise to a managed service with zero-downtime * Scalability & Performance: Shard rebalancing, customized scaling, and automated snapshots * Data Pipeline: 60TB of log data, and OpenTelemetry traces with Data Prepper * Operational Stability: Self-healing nodes and alert integrations with Slack. We will cover usage scenarios of OpenSearch dashboards and take a close look at how Aiven delivers and operates OpenSearch as a managed service at Nav’s scale.
CNCF Abassasor, Google Developer Expert (GDE) for Cloud, Grafana Champion and Platform Engineer at the Norwegian Labor and Welfare Administration (NAV) working on NAIS - a platform built to increase development speed by providing the best experience to build, run and operate appl... Read More →
Dmitry is a Product Director at Aiven, leading managed OpenSearch service. Since 2010, he has specialized in search (Lucene, Solr, Elasticsearch), previously co-owning TomTom’s planet-scale map search and building the technology behind AlphaSense. At Silo.AI, he led NLP and web-scale... Read More →
Most production monitoring follows the same path: metrics fire an alert, engineers search logs for context, and human reasoning becomes the bottleneck. A global p95 alert triggers, and 10–15 minutes later someone realizes the issue is limited to ap-south-1. Metrics and logs are tightly coupled through manual toil.
This session proposes a simple shift: logs detect first.
Structured JSON logs become system signals. These are indexed into purpose-built OpenSearch indexes, where sliding-window aggregation queries continuously evaluate system state. Queries like “p95 latency by region over 5 minutes” yield precise outcomes such as “REGION_OUTAGE: ap-south-1” instead of noisy global alerts.
Through a live demo, we show how explicit log schemas, index mappings, and Query DSL enable context-rich, machine-driven evaluation without dashboards or ad hoc searches. OpenSearch evaluates signals; actions remain external and decoupled.
The result is reduced human toil, a clearer system state, and a practical pattern that complements existing monitoring stacks without overpromising scale.
Shruti is a Cloud Engineer who builds bridges—from Git to GPU, from microservices to machine learning. A founding engineer turned DevOps specialist, she has architected multi-cloud systems on AWS and GCP, built self-service developer platforms with Terraform and ArgoCD, and instrumented... Read More →
Caching is essential for achieving low latency and high throughput in search systems, yet it’s often misunderstood or misused. OpenSearch and Lucene provide multiple caching layers—such as the request cache, field data cache, and Lucene query cache—each designed for different workloads and access patterns.
In this talk, we break down how OpenSearch caches work, when they help, and common pitfalls that lead to low hit rates or wasted memory. We share lessons from real production systems, showing how tuning, architectural choices, and benchmark-driven evaluation improve cache effectiveness, performance, and resiliency.
We then discuss optimizations across key cache layers — the request cache, field data cache, and Lucene query cache — and the impact they have on real workloads.
Finally, we introduce tiered caching, when it makes sense, how it integrates with existing cache layers, and the trade-offs involved.
This session is intended for engineers who want a practical understanding of caching in OpenSearch and Lucene and actionable guidance for improving search performance in production.
Software engineer at AWS Opensearch, Amazon Web Services
Sagar is a Software Engineer working on Amazon OpenSearch Service. He is an active contributor to the OpenSearch Project, primarily focused on search performance and resiliency-related features.
Ankit Jain is a Software Engineer on the Amazon OpenSearch Service team, leading performance and scalability initiatives for search infrastructure. He is an active maintainer and committer for the Apache Lucene and OpenSearch projects, with hands-on experience operating large-scale... Read More →
Friday April 17, 2026 16:00 - 16:40 CEST Bohemia 2
Kris Freedain (he/him) is the Senior Community Manager for the OpenSearch Project, OpenSearch Software Foundation, and technical steering committee. He is also an OpenSearch Ambassador. He has decades of experience in tech, but finds connecting people to be the most fulfilling part... Read More →
Patti Juric is Senior Marketing Content Manager for the OpenSearch Project, where she leads content, events, and partner strategy. She serves on the OpenSearch Software Foundation Marketing Committee and chairs the Program Committee for the Linux Foundation's Observability Summit... Read More →
Friday April 17, 2026 16:50 - 17:00 CEST Bohemia 2
