talk-data.com

We're excited to invite you to our next Meetup, co-hosted with the London Java Community!

Join fellow graph enthusiasts, Neo4j developers, and members of the Java community as we explore how graphs, knowledge graphs, and context engineering can unlock smarter applications and better answers. We’ll also uncover how AI-generated code can be weaponized, exposing hidden vulnerabilities in the software we trust. Whether you’re curious about building with graph-powered AI or safeguarding your projects from new threats, this meetup will give you insights you can put into practice.

Session 1: Black Friday Brilliance: Managing a Billion Transactions with Tech, Tactics, and Teamwork by Jamie Colmans

The Black Friday and Cyber Monday period is one of the busiest times in the retail calendar, both in stores and online, and here at Loqate our customers rely on our infrastructure to support their businesses at this crucial time. Over the four-day BFCM period in 2023 we processed over 1 billion requests to our APIs, and we managed this with greater than 99.99% availability! We've seen our request volume increase over 100x in the last 10 years, and managing this requires the right technologies, careful planning, and a great team of people.

With insightful commentary from a cross-section of our brilliant Dev team, I’ll talk you through how we scale our infrastructure to support these increases in traffic, and some of the technologies and processes we use. I’ll also give some insights into how the team works together over this busy period to keep everything running smoothly.

Session 2: WhatsThat? Using graphs and AI to make sense of your friends WhatsApp group by Martin O'Hanlon

Communication, messaging, and memory are complicated. How are you supposed to keep up with it all? It's difficult for you to keep up with the constant stream of messages you get. How can we expect an LLM to do the same? Let's look at how we can use graphs and AI to summarise the complex streams of messages you get in that oddly titled WhatsApp group.

We look forward to connecting with the community to exchange ideas on graphs, tech, and AI - join the discussion!

Dive into building applications that combine the power of Large Language Models (LLMs) with Neo4j knowledge graphs, Haystack, and Spring AI to deliver intelligent, data-driven recommendations and search outcomes. This book provides actionable insights and techniques to create scalable, robust solutions by leveraging the best-in-class frameworks and a real-world project-oriented approach. What this Book will help me do Understand how to use Neo4j to build knowledge graphs integrated with LLMs for enhanced data insights. Develop skills in creating intelligent search functionalities by combining Haystack and vector-based graph techniques. Learn to design and implement recommendation systems using LangChain4j and Spring AI frameworks. Acquire the ability to optimize graph data architectures for LLM-driven applications. Gain proficiency in deploying and managing applications on platforms like Google Cloud for scalability. Author(s) Ravindranatha Anthapu, a Principal Consultant at Neo4j, and Siddhant Agarwal, a Google Developer Expert in Generative AI, bring together their vast experience to offer practical implementations and cutting-edge techniques in this book. Their combined expertise in Neo4j, graph technology, and real-world AI applications makes them authoritative voices in the field. Who is it for? Designed for database developers and data scientists, this book caters to professionals aiming to leverage the transformational capabilities of knowledge graphs alongside LLMs. Readers should have a working knowledge of Python and Java as well as familiarity with Neo4j and the Cypher query language. If you're looking to enhance search or recommendation functionalities through state-of-the-art AI integrations, this book is for you.

Join us to explore the process of building knowledge graphs using Large Language Models (LLMs). This online session focuses on using LLMs to automatically extract knowledge graphs from unstructured data.

You'll learn how an LLM can identify key entities and their relationships, creating a structured representation of information. We will then demonstrate how to integrate this LLM-generated knowledge graph into a RAG pipeline, transforming it into a GraphRAG architecture for improved accuracy and contextual understanding in question answering.

This online session is tailored for developers interested in advancing their RAG implementations or other AI projects with the power of LLM-driven knowledge graphs.

Agenda

• Introduction / Meet the Community - 10 minutes
• Building Knowledge Graphs using LLMs - 40 mins
• Q&A - 10 mins

👉 New to SurrealDB? Get started here.

🗣️ Speaker opportunities - submit your talk! Working on an interesting project that you would like to share with the community? Submit your talk here.

FAQs

Am I guaranteed a ticket at this event? Our events are tech-focused and in the interest of keeping our events relevant and meaningful for those attending, tickets are issued at our discretion. We therefore reserve the right to refund ticket orders before the event and to request proof of identity and/or professional background upon entry.

Is this event for me? SurrealDB events are for software engineers, developers, architects, data scientists, data engineers, or any tech professionals keen to discover more about SurrealDB: a scalable multi-model database that allows users and developers to focus on building their applications with ease and speed.

Are there any House Rules? At SurrealDB, we are committed to providing live and online events that are safe and enjoyable for all attending. Please review our Code of Conduct and Privacy Policy for more information. It is compulsory for all attendees to be registered with a first and last name in order to attend. Any attendees who do not adhere to these requirements will be refused a ticket.

Join us for part two of Vector search and Graph use cases in SurrealDB! Learn how you can leverage this functionality in your own projects through informative talks with practical examples.

The meetup will highlight:

• The power of knowledge graphs in providing structured and semantically rich context to LLMs, leading to more informed and coherent responses.
• How vector embeddings, numerical representations of text that capture semantic meaning, enable semantic search within the knowledge graph, allowing the system to retrieve the most relevant information for a given query.
• A comparative demonstrating the difference in LLM responses when using:
• A standard prompt referencing source material alone.
• A prompt augmented by the knowledge graph and vector embeddings.

Attendees will gain practical insights into:

• The process of querying a graph-based RAG system for question answering.
• How to leverage the combined capabilities of SurrealDB, a multi-model database.
• Graph Capabilities: Representing relationships between entities within the knowledge graph.
• Vector Capabilities: Enabling semantic search to pinpoint relevant information within the knowledge graph.
• How this approach, utilizing SurrealDB's graph and vector features, enhances LLM responses by providing contextually relevant information retrieved through the knowledge graph.

This meetup is ideal for individuals who attended Part 1 or possess a basic understanding of knowledge graph extraction and are eager to learn advanced techniques for improving LLM outputs using graph-based RAG systems.

🗣️ Speaker opportunity - submit your talk! Working on an interesting project that you would like to share with the community? Submit your talk here.

⏰ Date/time: December 10, 6:30 - 9:00PM

📍 Location: The Yard: Columbus Circle Coworking Office Space NYC

Agenda

18:30 - 19:00 Welcome drinks, pizza & networking Attendees arrive – grab a drink, explore the space and meet the SurrealDB team.

19:00 - 19:30 Improving LLM Responses with Knowledge Graphs and Semantic Vector Search Sandro Pireno, Director Solutions Engineering at SurrrealDB. Building on the foundational concepts of knowledge graph construction from our last meetup in which we extracted knowledge graphs using a large language model (LLM),, this meetup explores advanced techniques for enhancing LLM outputs using graph-based Retrieval-Augmented Generation (RAG) systems. The session will showcase how integrating structured knowledge from a knowledge graph, coupled with semantic search powered by vector embeddings, can significantly improve the quality and relevance of LLM-generated responses.

19:30 - 20:00 Refreshments & networking Connect with others in the tech community. Grab a slice of pizza & a drink and chat with other attendees and members of the SurrealDB team.

20:00 - 20:30 How Index Uses SurrealDB with Decentralized Autonomous Agents Description: Explore how Index, a decentralized protocol for peer-to-peer discovery, integrates SurrealDB to enhance its network of autonomous agents. Discover how SurrealDB enables dynamic schemas, context-aware indexing, and seamless collaboration between agents.

20:30 - 21:00 Refreshments and networking

21:00 End of event

-- Host: Alessandro Pireno \| LinkedIn Alessandro is a seasoned product development and solutions leader with a proven track record of building and scaling data-driven solutions across diverse industries. He has led product strategy and development at companies like HUMAN and Omnicom Media Group, optimized data collection and distribution at GroupM, and was an early leader of success at Snowflake. With a deep understanding of the challenges and opportunities facing today’s tech landscape, Alessandro is passionate about empowering organizations to unlock the full potential of their data through innovative database solutions.

Guest speaker: Seref Yarar \| LinkedIn Seref Yarar is the co-founder of Index Network, with 15 years of experience across media, journalism, e-commerce, and ad-tech. His work is shaped by a focus on the semantic web, distributed systems, and decentralized technologies, which influence his approach to information discovery challenges.

--

👉 New to SurrealDB? Get started here.

FAQs

Is the venue accessible? The Yard is located on the 2nd floor. When you arrive, just let security know that you're heading up to The Yard.

Am I guaranteed a ticket at this event? Our events are tech-focused and in the interest of keeping our events relevant and meaningful for those attending, tickets are issued at our discretion. We therefore reserve the right to refund ticket orders before the event and to request proof of identity and/or professional background upon entry.

Is this event for me? SurrealDB events are for software engineers, developers, architects, data scientists, data engineers, or any tech professionals keen to discover more about SurrealDB: a scalable multi-model database that allows users and developers to focus on building their applications with ease and speed.

Are there any House Rules? At SurrealDB, we are committed to providing live and online events that are safe and enjoyable for all attending. Please review our Code of Conduct and Privacy Policy for more information. It is compulsory for all attendees to be registered with a first and last name in order to attend. Any attendees who do not adhere to these requirements will be refused a ticket.

We’re excited to announce our upcoming meetup in collaboration with Datenna, a pioneering scale-up based in Eindhoven. This event promises to be a deep dive into the innovative use of data and technology, showcasing cutting-edge applications that are shaping the future of open-source intelligence. And all this brought to you by the CTO and Founder of Datenna; Edward Brinkmann. In addition, Shu and Remi are sharing what they learned from building a tool to annotate LLM outputs. Sounds cool and interesting right? Datenna is our host this time and will open the doors of their office on October 29th, see you then!

How Datenna built a digital twin of China using graphs and GenAI How do you create a detailed and reliable digital twin of one of the largest economies in the world? How do you ensure that the data being collected from open sources is trustworthy? How do you handle conflicting pieces of information, merge entities across data sources, and ensure every conclusion is explainable and traceable back to the source? These are some of the challenges Datenna tackles daily in its mission to provide the best open-source intelligence to governments worldwide for economic and national security purposes. Discover how Datenna leverages graph databases and GenAI technology to build an open-source intelligence engine that continuously collects information on over 100 million entities in China, mapping all these entities and their relationships. Learn how Datenna, a scale-up founded in Eindhoven, has used these novel technologies to gain a competitive edge globally and became a world leader in techno-economic intelligence on China.

What we've learnt from building a tool to annotate LLM outputs LLMs can take files, audio, and video as input and generate summaries, answer questions, and extract information. With the Gemini family of models capable of supporting up to 1 million tokens in their context window, users can feed a PDF of hundreds of pages into these models and output only the results they care about. However, the outputs may contain errors. In this talk, the presenters will share their learnings from building a tool that enables the manual annotation and evaluation of these models' outputs based on a collection of models chosen by the users. They find the comparison results interesting and would like to share them with the audience.

Program

• 17:00 – 18:00 🍕 Food
• 18:00 – 18:10 🎤 Welcome
• 18:10 – 19:00 🎤 Edward Brinkman: How Datenna built a digital twin of China using graphs and GenAI
• 19:00 – 19:15 ⏸️ Break
• 19:15 – 20:00 🎤 Shu Zhao & Remi Baar: What we've learnt from building a tool to annotate LLM outputs
• 20:00-21:00 🥤 Drinks

About: Edward Brinkmann As the CTO and co-founder of Datenna, Edward has guided the company through various stages of growth, transforming it from a technology start-up to a thriving scale-up. His first role as CTO was as founding engineer, implementing the first versions of the intelligence platform, and later as engineering manager and lead architect whilst expanding the development team. With a background in software engineering, data engineering, and systems architecture, Edward has a broad interest in technology, especially in translating business needs into the most suitable technical solutions. Before co-founding Datenna, Edward enjoyed working on end-to-end projects in the capacity of lead developer and full-stack engineer, gaining experience across various business domains, use cases, and technologies.

About: Shu Zhao Shu has an MSc. in Artificial Intelligence, and part of her thesis was published in ECCV 2022 about artistic pose analysis based on computer vision. She also participated in AI Song Contest 2021 by producing one song by training an RNN-based language model.

Before she worked for Xebia data, Shu worked in various roles from large banks to smaller fintechs to e-commerce where she accumulated a wide span of technical skills to come to a sustainable solution.

About Remi Baar Fifteen years ago, at just 17, Remi launched his own software development company, quickly focusing on the exciting fields of artificial intelligence and data science. Since then, he has held various data science roles across a diverse range of organizations—from startups to multinational corporations, and from government agencies to airlines. His unique blend of software engineering expertise and data science has garnered him recognition and appreciation in each of these positions.

Currently, Remi is a valued member of the Xebia team, collaborating with fellow experts to enhance their collective skills and push the limits of AI. With a passion for knowledge sharing, Remi eagerly shares his latest insights.

Graph Retrieval Augmented Generation (Graph RAG) is emerging as a powerful addition to traditional vector search retrieval methods. Graphs are great at representing and storing heterogeneous and interconnected information in a structured manner, effortlessly capturing complex relationships and attributes across different data types. Using open weights LLMs removes the dependency on an external LLM provider while retaining complete control over the data flows and how the data is being shared and stored. In this talk, we construct and leverage the structured nature of graph databases, which organize data as nodes and relationships, to enhance the depth and contextuality of retrieved information to enhance RAG-based applications with open weights LLMs. We will show these capabilities with a demo.

RAG (Retrieval Augmented Generation) is a way to get LLMs to answer questions grounded in a particular knowledge base. What do you do when your knowledge base includes images, like graphs or photos? You first need to generate embeddings using a multimodal model, like the one available from Azure Computer Vision, search those embeddings using a powerful vector search like Azure AI Search, and then send any retrieved text and images to a multimodal LLM like GPT-4o. Learn how to get started quickly with a RAG on multimodal documents in this session.

Presented by Pamela Fox, Python Advocate at Microsoft

** Part of RAGHack, a free global hackathon to developer RAG applications. Join at https://aka.ms/raghack **

📌 Check out the RAGHack 2024 series here!

Pre-requisites: - Read the official rules and join the hack at https://aka.ms/raghack. No Purchase Necessary. Must be 18+ to enter. Contest ends 9/16/24.

• Want more hands-on RAG training? Visit the Reactor series home page to see all the RAGHack 2024 sessions!

Register: https://lu.ma/sakz1lmv

If you're passionate about AI, machine learning, data science, or linguistics, this event is for you. Connect with like-minded professionals, share insights, and learn from industry experts as they dive into the real-world applications of LLMs.

​Speakers & Topics: ​Lena Nahorna, Analytical Linguist at Grammarly Topic: Building Frameworks for Evaluation of LLM Output at Grammarly LLMs have opened up new avenues in NLP with their possible applications, but evaluating their output introduces a new set of challenges. In this talk, we discuss how the evaluation of LLMs differs from the evaluation of classic ML-based solutions and how we tackle the challenges.

​Halyna Oliinyk, Senior Data Engineer at Delivery Hero Topic: Data Engineering Workflow Before, After, and For LLMs Halyna will take you through the journey of deploying LLMs into production, focusing on the creation and management of modern data pipelines. She'll cover essential topics like system design, data sources, observability, and monitoring, all backed by real-world examples and common mistakes to avoid.

Djordje Benn-Maksimovic, Senior Data Scientist at Eviden Topic: Cypher Query Building with Open-Source LLMs Djordje will discuss creating knowledge graphs from news articles using small transformers for entity and relation extraction, and automating Cypher queries with open-source LLMs.

Perhaps the biggest complaint about generative AI is hallucination. If the text you want to generate involves facts, for example, a chatbot that answers questions, then hallucination is a problem. The solution to this is to make use of a technique called retrieval augmented generation, where you store facts in a vector database and retrieve the most appropriate ones to send to the large language model to help it give accurate responses. So, what goes into building vector databases and how do they improve LLM performance so much? Ram Sriharsha is currently the CTO at Pinecone. Before this role, he was the Director of Engineering at Pinecone and previously served as Vice President of Engineering at Splunk. He also worked as a Product Manager at Databricks. With a long history in the software development industry, Ram has held positions as an architect, lead product developer, and senior software engineer at various companies. Ram is also a long time contributor to Apache Spark.  In the episode, Richie and Ram explore common use-cases for vector databases, RAG in chatbots, steps to create a chatbot, static vs dynamic data, testing chatbot success, handling dynamic data, choosing language models, knowledge graphs, implementing vector databases, innovations in vector data bases, the future of LLMs and much more.  Links Mentioned in the Show: PineconeWebinar - Charting the Path: What the Future Holds for Generative AICourse - Vector Databases for Embeddings with PineconeRelated Episode: The Power of Vector Databases and Semantic Search with Elan Dekel, VP of Product at PineconeRewatch sessions from RADAR: AI Edition New to DataCamp? Learn on the go using the DataCamp mobile app Empower your business with world-class data and AI skills with DataCamp for business

How to get the most out of your enterprise knowledge graph, common challenges to be aware of, solutions and expected benefits & business cases.

Welcome to the first Connected Data London meetup of 2024! Doors will open at 6.00pm for talks starting at 18.30.

As usual refreshments and food will be available courtesy of our generous sponsors Beamery & Connected Data London.

Special perk for attendees: be the first to know the dates and venue for the upcoming Connected Data London 2024 conference!

Talk & Speakers

1. Knowledge Graphs x LLMs for Human Resources: automating the (once) impossible

Kaan Karakeben, Lead Data Scientist, Beamery

Much has been written about the ability of LLMs and Knowledge Graphs to work hand in hand. On the one hand you have Large Language Models that provide previously unimaginable flexibility and computational intelligence to automate any problem. Yet, without a KG to underpin this knowledge and ensure grounding, LLMs are subject to providing misinformation.

This talk will discuss Beamery’s approach to using LLMs and proprietary AI to “automate the hard stuff” in the HR profession, and how this work is underpinned by KGs.

2. Leveraging LLM and Knowledge Graphs for ESG Analysis

Adam Wangrat, Lead Knowledge Graph Engineer, Neural Alpha

This talk explores how integrating Large Language Models (LLMs) and Knowledge Graphs can revolutionize Environmental, Social, and Governance (ESG) research, disclosure analysis, custom benchmarking and other requirements.

By combining the power of an LLM-based question-answering system with a Neo4j Knowledge Graph containing rich ESG-related data, we can enable more accurate, context-aware, and insightful responses to user queries about ESG factors. Through real-world examples and practical strategies, attendees will learn how this synergy can transform ESG analysis, driving informed decision-making and stakeholder engagement.

3. How Graph RAG and rules-based AI can power expert chatbots and other applications

Nick Form, CTO, Oxford Semantic Technologies Cen Xi Toh, Knowledge Engineer, Oxford Semantic Technologies

This talk will show how to improve the accuracy of Large Language Models (LLM) by implementing a Retrieval Augmented Generation (RAG) approach using RDFox - a knowledge graph with an embedded semantic reasoner. We’ll demonstrate how this technology can power a chatbot to provide truthful, expert answers to even the most complex questions, using an example everyone can relate to - food!

The demo will concentrate on the use of an LLM in comprehending natural language, initiating function calls, and generating natural language responses. We’ll then cover how the use of rules and reasoning enhances the LLM's understanding and interpretation. The potential for other use cases across a variety of industries will also be explored.

Event timings 18:00 - 18:30 drinks, food & networking 18:30- 18:45 Welcome Beamery & Connected Data London 18:45- 19:15 First talk 19:15 - 19:45 Second talk 19:45 - 20:15 Third talk 20:20 - 21.00 Drinks/Networking 21.00 Event close

Address & building access Beamery, HYLO, 105 Bunhill Row, London EC1Y 8LZ

Your details will be shared with our event sponsor, Beamery, for communication, health & safety purposes relating to the event only.

You'll get an email with a QR code from Beamery Hylo building reception 4 - 24 hours before the event. Scan the code for entry when you arrive inside the building. If you don't receive the email, reception can print a code on the day if you've registered for a ticket or bring a plus one.

Who are Beamery? Beamery’s Talent Lifecycle Management platform brings together your data on candidates, employees and alumni.

We are an AI-powered platform that powers faster recruiting, successful internal mobility, smarter upskilling and more agile workforce planning so enterprise companies can improve the experience for all talent.

We are a team of creators, problem solvers and engineers. Here from the team directly at Beamery in this video about what life is like in the Engineering, Product & Design Team!

Hear more at 👉Careers.Beamery.com

Join us for an enlightening meetup where we dive into the fascinating world of graphs and generative AI.

Engage with like-minded individuals, share your experiences, and discover how Graphs and Artificial Intelligence can intersect to solve complex problems, enhance data visualization, and drive innovation across various industries. Whether you're looking to broaden your knowledge, apply these concepts in your work, or simply curious about the latest trends in technology, this meetup is the perfect platform to connect and learn.

Session: Using Knowledge Graphs in LLM applications Christopher Crosbie, VP GenAI, Neo4j

Discover how integrating Knowledge Graphs into the workflows of large language models enhances their understanding, reasoning, and response generation capabilities. The session covers various aspects, including the fundamentals of knowledge graphs and how they can provide context or factual backing to LLM outputs.

Join us for an engaging conversation about Large Language Models (LLMs) with a panel of industry leaders: Raja Iqbal (Chief Data Scientist, Data Science Dojo), Taimur Rashid (Chief Business Officer, Redis), Sam Partee (Principal Applied AI Engineer, Redis), and Daniel Svonava (CEO, Superlinked).

In this era where generative AI and LLMs are reshaping industries, businesses must be equipped to harness their potential fully. This discussion will focus on the common design patterns for LLM applications, especially the Retrieval-Augmented Generation (RAG) framework. The speakers will discuss various strategies for embedding knowledge into these models, using vector databases and knowledge graphs in fetching domain-specific data.

This discussion aims to not only inspire organizational leaders to reimagine their data strategies in the face of LLMs and generative AI but also to empower technical architects and engineers with practical insights and methodologies.

Join us for an engaging conversation about Large Language Models (LLMs) with a panel of industry leaders: Raja Iqbal (Chief Data Scientist, Data Science Dojo), Taimur Rashid (Chief Business Officer, Redis), Sam Partee (Principal Applied AI Engineer, Redis), and Daniel Svonava (CEO, Superlinked).

In this era where generative AI and LLMs are reshaping industries, businesses must be equipped to harness their potential fully. This discussion will focus on the common design patterns for LLM applications, especially the Retrieval-Augmented Generation (RAG) framework. The speakers will discuss various strategies for embedding knowledge into these models, using vector databases and knowledge graphs in fetching domain-specific data.

This discussion aims to not only inspire organizational leaders to reimagine their data strategies in the face of LLMs and generative AI but also to empower technical architects and engineers with practical insights and methodologies.

Join us for an engaging conversation about Large Language Models (LLMs) with a panel of industry leaders: Raja Iqbal (Chief Data Scientist, Data Science Dojo), Taimur Rashid (Chief Business Officer, Redis), Sam Partee (Principal Applied AI Engineer, Redis), and Daniel Svonava (CEO, Superlinked).

In this era where generative AI and LLMs are reshaping industries, businesses must be equipped to harness their potential fully. This discussion will focus on the common design patterns for LLM applications, especially the Retrieval-Augmented Generation (RAG) framework. The speakers will discuss various strategies for embedding knowledge into these models, using vector databases and knowledge graphs in fetching domain-specific data.

This discussion aims to not only inspire organizational leaders to reimagine their data strategies in the face of LLMs and generative AI but also to empower technical architects and engineers with practical insights and methodologies.