BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//papers.synesthesia.it//73KGHH
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-ai-heroes-2024-73KGHH@papers.synesthesia.it
DTSTART;TZID=CET:20241211T171500
DTEND;TZID=CET:20241211T180000
DESCRIPTION:We all learnt the basic modules of system design. Relational da
 tabase\, Blob storage\, API Gateway\, and Distributed Logging are familiar
  concepts. However\, the rise of Generative AI applications brings a new s
 et of challenges and tools to solve them.\n\nIn this talk\, we will explor
 e the new modules essential for developing Generative AI (GenAI) applicati
 ons\, addressing their unique challenges.\n\nWe will begin by talking abou
 t why using Large Language Models (LLMs) is insufficient. Next\, we'll int
 roduce guardrails\, which are crucial for sanitizing both the input and ou
 tput of LLMs and diffusion models. We'll then cover prompt compression tec
 hniques\, designed to reduce the cost of generation.\n\nFollowing this\, w
 e'll present prompt registries\, which enable over-the-air A/B testing and
  updates of prompts\, and the role of observability\, highlighting tools s
 uch as openllmetry\, a new standard based on opentelemetry.\n\nWe'll also 
 examine how to enhance LLM capabilities with function calls\, allowing LLM
 s to interact as agents and communicate with other external systems.\n\nDu
 ring the discussion\, we will show code examples in Python.\n\nBy the end 
 of this talk\, attendees will have a clear understanding of the main modul
 es used in building GenAI applications and gain valuable insights into des
 igning their own.
DTSTAMP:20241209T105218Z
LOCATION:Auditorium
SUMMARY:System Design for the GenAI Era - Emanuele Fabbiani
URL:https://papers.synesthesia.it/ai-heroes-2024/talk/73KGHH/
END:VEVENT
END:VCALENDAR