BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Callboard//Callboard Speaker Ops 1.0//EN CALSCALE:GREGORIAN METHOD:PUBLISH X-WR-CALNAME:Frontier AI Summit 2026 Schedule BEGIN:VEVENT UID:callboard-e6fc9e61-0333-403f-9940-90acf4db0ac8@demo.callboardhq.com DTSTAMP:20260813T073643Z DTSTART:20261007T150000Z DTEND:20261007T153000Z SUMMARY:R2 Lightning: Agents in Production Q&A LOCATION:Main Stage URL:https://demo.callboardhq.com/e/frontier-ai-summit-2026/schedule/e6fc9e6 1-0333-403f-9940-90acf4db0ac8 END:VEVENT BEGIN:VEVENT UID:callboard-000000sp-0000-4000-8000-000000000001@demo.callboardhq.com DTSTAMP:20260813T073643Z DTSTART:20261007T210000Z DTEND:20261007T213000Z SUMMARY:Building LLM Infrastructure That Scales: Lessons from Processing 10 0B+ Tokens Daily DESCRIPTION:Modern LLM applications demand infrastructure that handles mass ive scale reliably. We'll explore distributed serving patterns\, batching strategies\, and optimization techniques that work in production. Through real examples from systems processing billions of tokens daily\, you'll le arn how to architect inference platforms that maintain low latency while m aximizing throughput. We'll dive into containerization\, load balancing\, and graceful degradation patterns. By the end\, you'll understand the key tradeoffs in building production-grade LLM infrastructure and how to make architectural decisions that balance cost\, latency\, and reliability. Bri ng your current p99 and your monthly inference bill\; the capacity model w e walk through is the one that took ours from 940ms to 210ms without addin g a single GPU.\n\nSpeakers: Sam Speaker (Demo Speaker\, Callboard)\; Dr. Elena Rodriguez (Director of AI Infrastructure\, NeuroScale AI) LOCATION:Main Stage URL:https://demo.callboardhq.com/e/frontier-ai-summit-2026/schedule/000000s p-0000-4000-8000-000000000001 END:VEVENT BEGIN:VEVENT UID:callboard-000000sp-0000-4000-8000-000000000002@demo.callboardhq.com DTSTAMP:20260813T073643Z DTSTART:20261007T230000Z DTEND:20261008T003000Z SUMMARY:Workshop: Building Your First Agent System DESCRIPTION:Get hands-on experience building autonomous agent systems from scratch. We'll start with agent basics: state management\, decision loops\ , and tool integration. You'll build a working agent that can use multiple tools and handle errors gracefully. We'll explore different agent archite ctures: reactive agents\, planning agents\, and multi-step reasoning patte rns. Participants will code along with pre-configured environments. We'll discuss common failure modes and how to debug agent behavior. By the end\, you'll have a solid foundation for building agents and know where to dive deeper for production systems. Come with a laptop and a Python 3.11 envir onment. You will leave with a running agent\, a trace of every tool call i t made\, and a short list of the mistakes I made first so you can skip the m.\n\nSpeakers: Sam Speaker (Demo Speaker\, Callboard)\; Yuki Tanaka (Foun ding Engineer\, Agentic Systems Inc) LOCATION:Workshop Room 1 URL:https://demo.callboardhq.com/e/frontier-ai-summit-2026/schedule/000000s p-0000-4000-8000-000000000002 END:VEVENT BEGIN:VEVENT UID:callboard-000000sp-0000-4000-8000-000000000003@demo.callboardhq.com DTSTAMP:20260813T073643Z DTSTART:20261008T010000Z DTEND:20261008T013000Z SUMMARY:Evaluation Beyond Benchmarks: Building Evals That Catch Real Failur es DESCRIPTION:Standard benchmarks often miss the failures that matter in prod uction. This talk explores building comprehensive evaluation frameworks th at actually predict real-world performance. We'll discuss what makes evals brittle versus robust. You'll learn techniques for designing evals that a re rigorous without becoming expensive or gaming-prone. We'll cover automa ted evaluation systems\, human-in-the-loop approaches\, and continuous mon itoring. Practical examples from production systems show how different eva luation strategies catch different failure classes. Walk away with actiona ble patterns for building better evals. Fair warning: about half of this t alk is failures — three eval suites we shipped that stayed green while t he model got measurably worse\, and what we changed to make them able to f ail.\n\nSpeakers: Rina Okafor (Staff Engineer\, Vectorly)\; Marcus Chen (S enior Eval Engineer\, Clarity Evals) LOCATION:Workshop Room 2 URL:https://demo.callboardhq.com/e/frontier-ai-summit-2026/schedule/000000s p-0000-4000-8000-000000000003 END:VEVENT BEGIN:VEVENT UID:callboard-000000sp-0000-4000-8000-000000000004@demo.callboardhq.com DTSTAMP:20260813T073643Z DTSTART:20261008T210000Z DTEND:20261008T213000Z SUMMARY:Democratizing AI: Building Excellent Tools Without Massive Compute Budgets DESCRIPTION:You don't need billion-dollar budgets to build impactful AI too ls. This talk explores how solo founders and small teams are shipping comp etitive AI products through smart architecture choices. We'll discuss cost optimization strategies\, efficient model selection\, and architectural p atterns that maximize impact per dollar spent. Real examples show how open -source foundations and clever engineering solve problems previously requi ring massive resources. Learn about on-device inference\, caching strategi es\, and using models appropriately. Discover how constraint can drive inn ovation and how to build sustainable AI products as a bootstrapped founder . Aimed at solo builders and two-person teams. Every number in this talk c omes off a real invoice\, and the whole stack runs for less than the cost of a team lunch.\n\nSpeakers: Jamie Whitlock (Founder\, ML Developer\, Ind ieAI Labs) LOCATION:Main Stage URL:https://demo.callboardhq.com/e/frontier-ai-summit-2026/schedule/000000s p-0000-4000-8000-000000000004 END:VEVENT END:VCALENDAR