BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//hackathonmaxxing//EN
CALSCALE:GREGORIAN
X-WR-CALNAME:Hackathons in Paris
BEGIN:VEVENT
UID:luma:evt-0p69iiDnp1nz4pZ@hackathonmaxxing
DTSTAMP:20261122T163902Z
DTSTART:20261125T170000Z
DTEND:20261125T200000Z
SUMMARY:Daytona AI Builders - Paris\, November 2026
URL:https://lu.ma/ai-builders-paris-nov?utm_source=hackathonmaxxing&utm_medium=referral&utm_campaign=paris
LOCATION:Paris\, France
DESCRIPTION:An event dedicated to exploring all things AI Engineering!\n\nEvent partner: Replit\n\n​Agenda\n\n​​​​​🕒 6:00 pm – 6:30 pmGathering and Opening Remarks\n\n🎤 Juraj Stefanic\, Technical Content Engineer at Daytona\n\n​​​​​​​🕒 6:30 pm – 6:40 pmTalk "Give the Model a Computer: The One Thing Every AI Agent Framework Agrees On"​\n\n🎤 Mislav Ivanda\, DeEx Engineer at Daytona\n\n​​​​​​​​Outline:\n\nBuilding an agent is easy. Deciding where its code runs is the part that follows you to production.\n\nThis talk comes from a year of building Daytona integrations across the agent ecosystem: LangChain\, CrewAI\, Google ADK\, Mastra\, Pydantic AI\, the OpenAI Agents SDK\, and a long tail behind them. Integration after integration\, the same pattern kept surfacing. Frameworks that agree on nothing\, not even the language\, not the mental model\, not who controls the flow\, all converge on one identical primitive: give the model a computer. We'll trace why.\n\nEvery framework ends up facing the same question: where does the model's code actually run? Only three answers exist\, and two of them stop holding once real users show up. That's why the past year went the way it did\, with frameworks deleting their own code interpreters rather than defending defaults they couldn't. And that's why the computer keeps growing: first a tool call\, then the place the agent lives\, then a thousand machines at once.\n\nYou'll leave with one rule: pick the boundary before the framework. Frameworks are opinions. The boundary is a commitment.\n\n​🕒 6:40 pm – 6:50 pmTalk "Faster\, Safer Agentic Workloads with NVIDIA Vera and OpenShell"\n\n🎤 Asma Farjallah\, Senior AI Developer Technology Engineer at NVIDIA\n\n​​​​​​​Outline:\n\nAsynchronous RL post-training overlaps rollout generation with policy updates\, cutting GPU idle time at the cost of some off-policy staleness. But agentic rollouts add CPU- and network-bound steps whose latency can still bottleneck GPU utilization. We walk through the asynchronous post-training loop and highlight the CPU-boun https://lu.ma/ai-builders-paris-nov?utm_source=hackathonmaxxing&utm_medium=referral&utm_campaign=paris
END:VEVENT
END:VCALENDAR
