Brainberg
Building Authorization for Agentic RAG Systems
AI Integration & ApplicationMeetupFree

Building Authorization for Agentic RAG Systems

Wed 2 Dec · 16:45
Amsterdam, 🇳🇱 Netherlands
< 50 attendees
Elasticsearch, Inc. · Keizersgracht 281

About this event

Join us for a meetup together with AuthZed at the Elastic office in Amsterdam.

Agenda:
17.45: Doors open
18.00: Welcome
18.05: Creating Custom Kibana Dashboards with Elastic Inference Service
18.50: Building Authorization for Agentic RAG Systems
19.30: Networking, pizza and drinks
20.30: Wrap up

Talks:
Creating Custom Kibana Dashboards with Elastic Inference Service

Kibana dashboards are easy to build, but custom reporting still takes time. You need to know the data, the fields and the query language. AI-assisted generation changes this. You describe the report you want, and Elastic's own inference service builds the content from the data already in your cluster.

That speed creates a risk. If the assistant can query anything the user's role can reach, a developer can end up asking questions about finance data, or a sales rep can summarise support tickets. The generated report then turns that access into a polished, shareable summary, so a gap in role design becomes much more visible.

You will see how to go from raw cluster data to a generated dashboard and report. You will leave with a short checklist to run before enabling AI-generated reporting: scope roles to the data they need, separate sensitive indices, and check what each role can actually retrieve.

Speaker: Afzaal Ahmad Zeeshan, Developer Advocacte @ Elastic

Building Authorization for Agentic RAG Systems
We’ve recently seen the rise of Retrieval Augmented Generation (RAG) systems. With the proliferation of AI Agents & LLMs in Enterprise, the next step in its evolution is Agentic RAG. However, widespread enterprise adoption requires solving the puzzle of fine-grained authorization.

This talk is a deep dive into how modern authorization systems can ensure that AI Agents have access only to authorized data. The talk will look at why the Google Zanzibar model of authorization which uses Relationship-Based Access Control (ReBAC) is well suited to handle dynamic, relationship-driven authorization at scale. The talk explains how the Google Zanzibar system works under the hood and how to apply it to Agentic RAG with techniques such as pre-filtering and post-filtering.

The talk will also include a live coding demo implementing authorization for Agentic RAG using Open Source tools such as Elasticsearch, Langchain, and SpiceDB.

Speaker: Sohan Maheshwar, Lead Developer Advocate @AuthZed

Source: meetup