BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Superpower Daily//AI Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Superpower Daily AI events
BEGIN:VEVENT
UID:b4cf627e-938c-4542-bdca-ee2f1f5c25ca@superpowerdaily.com
DTSTAMP:20261011T200356Z
LAST-MODIFIED:20261011T200356Z
DTSTART:20261023T010000Z
DTEND:20261023T040000Z
SUMMARY:Training Agents IRL
DESCRIPTION:GitHub and Hugging Face will bring together ML practitioners an
 d RL researchers for an evening diving into how AI agents are trained and 
 evaluated—covering everything from reward design and multi-step training
  to the real-world gaps that matter. Expect rapid-fire talks from industry
  builders at Microsoft\, Hugging Face\, and emerging agent startups\, plus
  a panel unpacking what actually works in agent RL and where the ecosystem
  needs to go next.\nDetails: https://superpowerdaily.com/events/training-a
 gents-irl-b4cf627e-938c-4542-bdca-ee2f1f5c25ca\nRegister: https://luma.com
 /awautg3p
LOCATION:San Francisco\, CA
URL:https://superpowerdaily.com/events/training-agents-irl-b4cf627e-938c-45
 42-bdca-ee2f1f5c25ca
STATUS:TENTATIVE
TRANSP:TRANSPARENT
END:VEVENT
END:VCALENDAR