Tracing Agent Decisions with Graph Evals and Neo4j

AI agents don’t just need to perform, they need to be understood, trusted, and improved.
Traditional evals only look at inputs and outputs, ignoring the messy middle where most agent failures happen.

This session introduces Graph Evals, a practical technique where every agent step (actions, states, tool calls, reasoning hops, failure points) is stored as a knowledge graph. By modeling an agent’s internal decision journey in Neo4j, we can analyze its reasoning patterns, detect blindspots, identify loops, and understand why it behaved the way it did.

Attendees will learn how to build a graph-based eval pipeline, visualize agent reasoning paths, run structural queries to catch failure modes, and continuously refine agent policies using graph insights.

Perfect for teams deploying production-grade agents and anyone who wants their AI to act less like a black box.

Beyond Text - A Practical Guide to Multi-Modal Knowledge Graph Pipelines

Real-world data isn’t just text. PDFs, spreadsheets, images, screenshots, charts, and videos all contain valuable knowledge, but extracting and structuring it is hard.

This session walks through a practical, end-to-end pipeline for ingesting multi-modal data (text, tables, charts, images, drawings) into your Knowledge Graph using Neo4j.

Learn how to extract, enrich, and link diverse content into a unified graph ready for GraphRAG, agents, and reasoning.

Make Your AI Think - Adding a Reasoning Layer to Your RAG Stack with Neo4j

LLMs are powerful, but retrieval alone only answers what. In real projects, stakeholders need to know why things happened, who was responsible, and how entities connect across years of scattered documents.

This session shares a real estate case study where 15+ years of contracts, PDFs, and project files were transformed into a knowledge graph. By combining vector retrieval with Neo4j’s reasoning, we moved beyond semantic matches to multi-hop logic and explainable insights.

Learn how to build a reasoning layer for your RAG stack that grounds answers, reduces hallucinations, and delivers business intelligence your stakeholders can trust.

Prompt, Plan, Perform - The Agentic Revolution in AI

Curious about how AI really works under the hood? Tired of hearing buzzwords like “LLM” and “Agents” thrown around without context? This session is your no-fluff, high-energy, beginner-friendly guide to understanding the most disruptive tech of the decade, Large Language Models and AI Agents.

From prompting to planning, you’ll learn how AI agents can act, reason, and operate with goals, making them useful for tasks far beyond simple chatbot Q&A.

Key learnings
- Clearly define what a Large Language Model (LLM) is and how it works
- Understand the strengths, limitations, and ideal use cases for LLMs
- Learn what differentiates AI Agents from chatbots and traditional tools
- Walk away with a framework to evaluate and build LLM-powered solutions

Beyond Retrieval: Adding a Reasoning Layer to RAG with Neo4j

The reasoning layer in my RAG system is powered by Neo4j’s knowledge graph, which acts as a structured context source beyond flat document retrieval.

This hybrid system lets the LLM “think” in terms of relationships, not just keywords, adding explainability and depth to its answers.

Building Knowledge Base for Real Estate using Gemma and Neo4j

In this presentation, we delve into the development and deployment of a custom Large Language Model (LLM) tailored for a major infrastructure company. Leveraging the capabilities of Gemma and Neo4j, we created an advanced, queryable knowledge base that consolidates data from past real estate projects. This solution not only enhances data accessibility and retrieval but also demonstrates the transformative potential of AI in the real estate sector. Attendees will gain insights into the design, implementation, and benefits of integrating LLMs with robust database systems to streamline knowledge management and drive informed decision making.

A Deep Dive into Design System

In today's rapidly evolving tech world, designing digital products that are user-friendly, visually appealing, and scalable has become increasingly challenging. To meet these challenges, many companies are adopting design systems, a collection of reusable components, guidelines, and best practices that help create consistent and cohesive user experiences across products and platforms.

In this session, we will take a deep dive into design systems, discussing the key concepts, benefits, and challenges of adopting a design system for your organization. We will explore the different types of design systems, including style guides, component libraries, and pattern libraries, and how they can be customized to fit your specific needs.

We will also look at real-world examples of design systems from industry leaders like Google, Airbnb, and Salesforce, and examine their design principles and best practices. You will learn how to create and maintain a design system, how to ensure its adoption and success within your organization, and how to measure its impact on your design and development processes.

Whether you are a designer, developer, or product manager, this session will provide valuable insights into how design systems can improve your team's productivity, consistency, and creativity. Join us to learn from the experts and take your design skills to the next level.

A brief history of JavaScript bundlers

JavaScript bundling has been a cornerstone of web development, and over time, bundling tools have undergone significant transformations. In this talk, we'll delve into the fascinating history and evolution of JavaScript bundlers.

We'll explore the inner workings of these tools, uncovering the fundamental problems they were designed to solve. Discover how bundlers have adapted to the changing landscape of web development and how their evolution impacts developers and applications today.

Whether you're a seasoned developer interested in the history and intricacies of your favorite tools or a newcomer looking to understand the pivotal role bundlers play in modern web development, this session will provide valuable insights. Join us to gain a deeper appreciation for the evolution of JavaScript bundlers and how it shapes the way we build web applications.

A practical guide to GraphQL

Join us for an enlightening journey into the world of GraphQL as we demystify this revolutionary technology. In this engaging talk, you'll gain a deep understanding of GraphQL's power and versatility. We'll break down the complexities into practical, actionable insights, making them accessible to newcomers and experienced developers alike.

Discover how GraphQL simplifies data querying, empowers front-end developers, and streamlines your API architecture. You'll learn how to design efficient, flexible APIs and avoid common pitfalls through real-world examples and hands-on demonstrations. From building queries to optimizing performance, we'll provide a comprehensive guide for harnessing the true potential of GraphQL.

Leave this session equipped with the knowledge and tools you need to implement GraphQL in your projects, enhancing your development capabilities and setting you on the path to creating more efficient, adaptable, and responsive applications. Whether you're a developer looking to level up your skills or a tech enthusiast eager to explore the future of data querying, this practical guide is your key to unlocking the full potential of GraphQL.

Discover how GraphQL simplifies data querying, empowers front-end developers, and streamlines your API architecture. You'll learn how to design efficient, flexible APIs and avoid common pitfalls through real-world examples and hands-on demonstrations. From building queries to optimizing performance, we'll provide a comprehensive guide for harnessing the true potential of GraphQL.

Retrieval Augmented Generation (RAG) explained with Gemma

The session explains various approaches that help LLMs to create smarter, more accurate AI systems. We will breakdown how RAG works, from pulling relevant information from large databases to generating nuanced, context-rich responses. We will also explore some real world use cases along with a small demo using Google Gemma. Whether you’re new to AI or looking to deepen your knowledge, this session will show you how RAG is shaping the future of intelligent systems.

Beyond Tables - Leveraging Graph Database for Modern Data Challenges

Explores the limitations of traditional relational databases in handling complex, interconnected data. The session highlights how graph databases excel in scenarios like social networks, fraud detection, recommendation systems, and supply chain management. It covers the fundamentals of graph databases, their benefits, and real-world use cases, empowering participants to solve modern data challenges with graph-powered insights.