BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Palaner//palaner Single Event//EN
METHOD:PUBLISH
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:meetup-7e73c51d20@palaner.app
DTSTAMP:20260802T221224Z
DTSTART:20260804T173000Z
DTEND:20260804T190000Z
SUMMARY:PyData London - 109th Meetup
LOCATION:EC4R 3AD
DESCRIPTION:Saved in Palaner as: saved\n\nPalaner match: 37% · why: lots of
  people going\n\nVenue: Riverbank House\, 2 Swan Ln\, London EC4R 3AD\nPle
 ase note:\n1. 🚨🚨🚨 A valid photo ID is required by building security. 🚨🚨🚨\n
 2. This event follows the NumFOCUS Code of Conduct. Please familiarise you
 rself with it before attending.\n\nIf your RSVP status says "You're going"
  you will be able to get in. No need to show your RSVP confirmation when s
 igning in.\nIf you can no longer make it\, please unRSVP as soon as possib
 le.\n\nCode of Conduct:\nThis event follows the NumFOCUS Code of Conduct. 
 Please get in touch with the organisers with any questions or concerns.\n\
 nAs always\, there will be free food and drinks\, generously provided by o
 ur host\, Man Group.\n\nMain Talks\n\n1. Sefik Serengil - RAG Is Dead\, Lo
 ng Live GraphRAG!\nTraditional RAG pipelines typically break documents int
 o independent chunks—such as paragraphs or pages—and convert each chunk in
 to an embedding. While effective for basic retrieval\, this approach loses
  the semantic continuity that naturally exists across chunk boundaries. A 
 sentence at the end of one page may be tightly connected to the first sent
 ence of the next\, yet classic RAG treats them as unrelated vector islands
 . Neo4j’s open-source GraphRAG project addresses this fundamental limitati
 on by converting chunks into a graph representation where entities\, relat
 ionships\, and contextual links are explicitly modeled. Instead of isolate
 d embeddings\, GraphRAG creates a connected structure in which both the ch
 unks themselves and the entities inside them can be embedded using graph a
 lgorithms and LLM-based backends. This allows semantic meaning to flow acr
 oss the graph\, preserving context at multiple levels while enabling riche
 r retrieval and more accurate reasoning.\nIn this talk\, we will explore h
 ow GraphRAG builds these graph-native representations\, how it leverages N
 eo4j to store and traverse them\, and how its embedding strategies differ 
 from conventional RAG systems. We will also discuss practical workflows\, 
 example architectures\, and the benefits of using a graph-oriented approac
 h for tasks that require higher-fidelity context understanding. Attendees 
 will gain a clear picture of why GraphRAG matters\, what problems it solve
 s\, and how to start using the open-source package in real-world applicati
 ons.\n\nLightning Talks\n\n1. Rafael Garcia-Dias - AI on the NHS: Can fede
 rated learning preserve patient privacy?\n2. Chaerin Kwon - The Hardest Wo
 rd to Translate: Teaching Machine Translation to Say "You" in Korean\n3. P
 eter Bleackley - Emily - An Open Source Information Retrieval System\n\n--
 ------------------------------\n\nLogistics\n\nDoors open at 6.30 pm (get 
 there early as you'll need to sign in with building security).\nTalks star
 t at 7:00 pm\, with drinks afterwards from 9:00 pm at The Banker (EC4).\n\
 nWe have reduced capacity for this event\, but there will be plenty of peo
 ple to discuss data science questions with.\nPlease unRSVP in good time if
  you realise you can't make it. We're limited by building security on the 
 number of attendees\, so please free up your place for your fellow communi
 ty members.
URL:https://www.meetup.com/pydata-london-meetup/events/315916503/
CATEGORIES:palaner,saved
STATUS:TENTATIVE
END:VEVENT
END:VCALENDAR