BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.pydata.org//pydataglobal2025//speaker//7DWP9K
BEGIN:VEVENT
UID:pretalx-pydataglobal2025-7PTYQX@cfp.pydata.org
DTSTART:20251211T183000Z
DTEND:20251211T190000Z
DESCRIPTION:Understanding customer behavior is essential in marketing. Trad
 itionally\, marketers rely on methods such as surveys\, customer interview
 s\, and focus groups to gather insights. However\, these approaches can be
  expensive\, time-consuming\, and limited in scale and diversity.\nRecentl
 y\, multi-agent simulation powered by Large Language Models (LLMs) is emer
 ging as an innovative technique.  TinyTroupe\, for example\, enables the c
 reation of different personas (e.g.\, budget‑minded Gen‑Z shoppers\, p
 remium‑seeking parents)\, allowing marketers to predict and optimize adv
 ertising effectiveness or replace time-consuming interviews rapidly.\nIn t
 his talk\, I will introduce the key concepts of LLM-powered multi-agent si
 mulations\, demonstrate their practical application in marketing through T
 inyTroupe\, and share actionable insights and recommendations.
DTSTAMP:20260718T235308Z
LOCATION:Machine Learning & AI
SUMMARY:TinyTroupe: Enhancing Marketing Insights through LLM-Powered Multia
 gent Persona Simulation - Hajime Takeda
URL:https://cfp.pydata.org/pydataglobal2025/talk/7PTYQX/
END:VEVENT
END:VCALENDAR