BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pydata-london-2026//talk//HAYANG
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pydata-london-2026-HAYANG@pretalx.com
DTSTART;TZID=GMT:20260607T153000
DTEND;TZID=GMT:20260607T161500
DESCRIPTION:Large Language Models are rapidly changing how we think about r
 ecommendation systems. Traditional pipelines based on collaborative filter
 ing or matrix factorization are being complemented and sometimes replaced 
 by embedding-based and LLM-driven approaches.\n\nIn this talk\, we explore
  how modern recommendation systems can be built using LLM embeddings\, vec
 tor databases\, and hybrid architectures that combine classical ML with ge
 nerative models. We will discuss practical design patterns for personaliza
 tion\, retrieval\, ranking\, and user modeling\, focusing on real-world co
 nstraints such as latency\, cost\, and evaluation.\n\nThe session emphasiz
 es hands-on insights from production systems and highlights where LLMs add
  real value and where they don’t. Attendees will leave with a clear ment
 al model for designing scalable\, LLM-powered recommendation systems beyon
 d toy examples.
DTSTAMP:20260602T223426Z
LOCATION:Grand Hall 1
SUMMARY:LLM-Based Recommendation Systems: From Embeddings to Real Personali
 zation - Özge Çinko
URL:https://pretalx.com/pydata-london-2026/talk/HAYANG/
END:VEVENT
END:VCALENDAR