The next Elastic UK Public Sector Summit will help you turn today's data risks into stronger, more resilient public service outcomes.
Practical strategies to stay ahead of data challenges
Across government, healthcare, defence, higher education and public safety, organisations are facing growing pressure to secure critical data, harness AI responsibly and make faster, more informed decisions. Join peers and experts to discover practical strategies that help you stay ahead of tomorrow's data challenges.
We'll bring together public sector leaders, practitioners and Elastic experts to explore today's biggest data challenges while preparing for tomorrow's - from cyber resilience and AI governance to operational efficiency and citizen outcomes.
You'll hear from high-calibre keynote speakers from the highest levels of government, NHS, defence, higher education and public safety, alongside interactive, mission-focused breakout sessions with actionable insights.
With a high-profile lineup and limited capacity, places will be allocated on a first-come, first-served basis to delegates from public sector organisations.
• Hear from Public Sector Leaders: Connect with peers who are solving real data, AI and security challenges across the UK Public Sector.
• Agenda Built Around Your Challenges: Designed from public sector feedback, every session focuses on real-world challenges, practical solutions and measurable outcomes.
• Strategic Ecosystem: Engage with leading technology partners and consultants delivering secure, government-ready solutions.
South London and Maudsley NHS Foundation Trust (SLaM) | Dr. David Codling, Clinical Lead for the Clinical Informatics Service
Text data is hugely important in healthcare, particularly mental healthcare, containing valuable insights into diagnosis, risk factors and treatment. However, we need to be clear when aggregating this data that the underlying text is surfaced and searchable by clinicians.
This talk describes the journey to design a safe, effective system for digitally transforming how community mental healthcare is delivered using Kibana at South London and Maudsley Trust.
250+ systems. A mountain of logs. What happens when you start joining the dots?
UKHO’s observability journey began by bringing logs, performance telemetry, synthetics and traces together to understand our services more clearly. But seeing what’s happening is only the start. We’re now using SLOs, operational insight and AI-assisted diagnosis and remediation to turn visibility into action, while expanding coverage and keeping a close eye on cost.
We’ll share what we’ve learnt, what’s working, and some practical things you can take away and try tomorrow.
Next, we will invite you to choose from the following 25-minute technical deep dive sessions:
Track 1: Defensible AI
The AI audit trail: every prompt, every source.
AI is already shaping casework, triage and operations. See how to capture every prompt, retrieved document and response as a searchable, access-controlled record, so you can account for it to an auditor years later.
Sovereign AI you can observe: grounded and measured
Run inference inside your own boundary, ground answers in your documents with citations, and observe the AI itself: tracing LLM calls, flagging ungrounded answers and tracking token spend.
Context Engine: why your agents keep getting it wrong, and what to feed them instead
Most agent failures are context failures. See how mapping your sources ahead of time and serving a precise slice, with provenance and your access controls enforced, makes agents cheaper, more accurate and accountable.
Track 2: Operations and security, human accountability
Agentic security operations: analysts who direct decisions, not drown in alerts
Start with AlertZero, then see its Watches triage, investigate and group alerts into incidents, draft the response and hand the decision to an analyst: more of the estate watched by the same number of people.
Agentic SRE: from alert to answer in one conversation
A citizen-facing service goes down at 9am on a Monday. See AI-assisted investigation correlate logs, metrics and traces, propose a likely root cause and explain it in plain language to the person who briefs the minister.
Break it before they do: purple teaming your detections
Stand up disposable, realistic ranges, run known adversary techniques through them and measure which detections fire and which do not. Then fix the gaps as routine engineering, not an annual exercise.
Elastic Defend: Rethinking Endpoint Security
A straight answer to “can’t we just consolidate?”: what Elastic Defend does well, what it deliberately does not do, and how it fits alongside the management tools you keep.
Track 3: Insight where the data lives
Elastic 101: one platform for search, security and observability, wherever your data has to live
The starting point for anyone new to the platform: how one platform covers search, security and observability, and how the same stack runs in the cloud, on-premises or fully air-gapped.
One question across estates, without moving the data
Query across separate clusters, departments and classification boundaries in a single request, with the data staying exactly where governance says it must.
Standards first: observability across a fragmented estate
Build one observability capability across a large, fragmented organisation by standardising on OpenTelemetry first, so each team keeps its own tools while the organisation gets one view of logs, metrics and traces.
Elasticsearch is a trademark of Elasticsearch B.V., registered in the U.S. and in other countries. Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS, and the yellow elephant logo are trademarks of the Apache Software Foundation in the United States and/or other countries
