BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//events//Events//EN
CALSCALE:GREGORIAN
X-WR-CALNAME:Building AI Memory: From a Local Robot to a 1\,000-Fly Hive Mi
 nd
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
BEGIN:VEVENT
UID:event-auz233i5@ontown.app
DTSTAMP:20261110T091753Z
DTSTART:20261117T170000Z
SUMMARY:Building AI Memory: From a Local Robot to a 1\,000-Fly Hive Mind
LOCATION:San Francisco\, United States
DESCRIPTION:ODSC AI San Francisco\n​Pre-Registration is REQUIRED. Joining
  link -\nhttps://streamyard.com/watch/d5uJc3J3eSbe (https://streamyard.com
 /watch/d5uJc3J3eSbe?utm_source=luma)\n\nHow do you give a fully local robo
 t the ability to remember what it sees? How do you let 1\,000 flies share 
 what they learn?\n​Join Dylan Couzon from Qdrant for a free\, live webin
 ar hosted by ODSC AI\, supported by Qdrant.\n\n​WHEN\n​Tuesday\, Novem
 ber 17\, 2026 | 12:00 PM ET / 9:00 AM PT\nOnline · Free with registration
 \n​ABOUT THE SESSION\n​We walk through building both memory systems\, 
 from storing the first observation to retrieving useful experiences and sh
 aring them across agents. Along the way\, we tackle the decisions that mak
 e memory useful: what to keep\, when to retrieve it\, what to share\, and 
 when to forget.\n​See the systems run live\, inspect the code behind the
 m\, and learn patterns you can apply to your own AI agents.\n​WHAT YOU
 ’LL LEARN\n\n• ​Why more memory can make agents worse\n• ​How to
  design what gets written\, retrieved\, and shared\n• ​How to scope sh
 ared memory across many agents\n• ​What changes when memory runs fully
  on-device\n\n​WHO SHOULD JOIN?\n\n• ​AI engineers\, software develo
 pers\, and data scientists\n• ​People building or experimenting with A
 I agents\, retrieval systems\, or robotics\n• ​Product managers and te
 chnical leaders exploring how agent memory works\n\n​MEET THE SPEAKER\n
 ​Dylan Couzon - Lead Developer Relations\, Qdrant\n​Dylan leads Develo
 per Relations at Qdrant. He learned to code 18 years ago to work on video 
 game emulators and has been taking side projects too far ever since. He ha
 s spent his career building developer tools and contributing to open sourc
 e. These days\, he builds retrieval systems\, trains models\, and experime
 nts with physical AI and robotics.\n\n​Useful Links\n\n• ​*\\odsc.ai
  (https://odsc.ai/?utm_source=luma)\\*\n• ​Free access to more talks/t
 rainings: Ai+ Training platform (https://aiplus.training/?utm_source=luma)
 \n• ​ODSC Blog: opendatascience.com (https://opendatascience.com/?utm_
 source=luma)\n• ​Slack Channel: Join the Community (https://odsc.ai/sl
 ack?utm_source=luma)\n\n!User Uploaded Image (https://images.lumacdn.com/c
 dn-cgi/image/format=auto\,fit=cover\,dpr=2\,quality=75\,width=800\,height=
 108.843537414966/uploads/sw/a23d89ac-0849-4a14-b6d3-a8991bb6f6f8.png)\n!Us
 er Uploaded Image (https://images.lumacdn.com/cdn-cgi/image/format=auto\,f
 it=cover\,dpr=2\,quality=75\,width=800\,height=256/uploads/hg/262d9b89-4e0
 6-4574-80a2-800a149a810e.png)\n​We're the ultimate builder-first communi
 ty for data scientists\, AI/data engineers\, and developers. We’re movin
 g past the pitch decks to bring the real meetup vibes back. This is all ab
 out raw live demos\, discovering new tools\, hacking with AI agents\, and 
 getting the inside scoop from the people who actually ship AI. Whether you
 ’re here to watch an agentic workflow layout live\, crack open some fres
 h open-source frameworks\, or just here for the pizza\, cold beers\, and a
  genuinely good time with real humans\, we’ve got your seat saved.
URL:https://ontown.app/e/auz233i5-building-ai-memory-from-a-local-robot-to-
 a-1000-fly-hive-mind/
END:VEVENT
END:VCALENDAR