For developers: this programme is open data — JSON, iCal, schedule XML and an MCP endpoint.Show endpointsHide
- JSONEvery published session and speaker, in one request./aie-worldsfair-2026-import/feed.json
- iCalSubscribe in Google, Apple or Outlook Calendar./aie-worldsfair-2026-import/feed.ics
- Schedule XMLfrab / pentabarf — the format conference apps import./aie-worldsfair-2026-import/feed.xml
- MCP + RESTPoint Claude at the programme. OpenAPI 3.1 included./agents
No key, no signup, CORS open. Everything here is generated from the same data the organisers edit.
Showing 561 of 561 sessions · 4 days · 50 tracks
Monday, June 29
58 sessions9:00 AM · Track 1From Vibes to Production: Evaluating and Shipping AI Agents That Work 101Laurie VossTrack 19:00 AM · Track 2AI on Your Lakehouse: Context Comes in Shapes, Not QueriesZach BlumenfeldTrack 29:00 AM · Track 3Cooking with CodexCharlie Guo, Gabriel ChuaWorkshops Day 19:00 AM · Track 4The best SDLC is the one you build yourself: Why orchestration changes everythingShane Wolf, Andrei BocanWorkshops Day 19:00 AM · Track 5AI Security Engineer Foundations + CertificateJavier GarzaWorkshops Day 19:00 AM · Track 6Total Recall: Agent Memory and Harness EngineeringIgnacio MartinezWorkshops Day 19:00 AM · Track 7Agents That Own Their Inference: Building Production AI Agents on Dedicated GPUsDu'an Lightfoot9:00 AM · Track 8Open-Source Inference Engineering for the Agentic EraZain Hasan, Yubo Wang, Qingyang Wu, Jue WangWorkshops Day 19:00 AM · Track 9Advanced workshop: Mastering AI ObservabilityDoug Guthrie9:00 AM · Track MGet Started with Models in Microsoft Foundry to Build AI AppsPamela FoxTrack M11:05 AM · Track 1Building self-learning loops for your agentFuad AliPosttraining & Midtraining11:05 AM · Track 2RAG Needs a Map: Using GraphRAG to Retrieve Connected ContextNyah MacklinTrack 211:05 AM · Track 3How I learned to stop worrying and love the sandboxMatt BrockmanWorkshops Day 111:05 AM · Track 4The model swap workshopPamela Fox, Arun SekharWorkshops Day 111:05 AM · Track 5Teaching Agents to Search: Building Synthetic Training Pipelines with NVIDIA Data DesignerDhruv NathawaniWorkshops Day 111:05 AM · Track 6Local LLMs and workstation agents: Part 1Ahmad OsmanWorkshops Day 111:05 AM · Track 7How to Build Quality Gates into Agentic Coding WorkflowsNnenna NdukweWorkshops Day 111:05 AM · Track 8What is an Inference Engine, Anyway?Charles FryeWorkshops Day 111:05 AM · Track 9Agent Speedrun: Idea → Code → Deploy → Observe, Fix → ShipElizabeth Fuentes Leone, Sandhya SubramaniWorkshops Day 111:05 AM · Track MFrom zero to deployed on Azure with AI agentsGustavo CordidoTrack M12:10 PM · Track 1Evals in AI: A Deep DiveTejas KumarWorkshops Day 112:10 PM · Track 2From approval loops to autonomous agents with DockerJohn Craft, Dan NdombeWorkshops Day 112:10 PM · Track 32 hr deep dive on LLM Inference at Scale — Part 1 of 2Harshul Jain, Tanmay SahWorkshops Day 112:10 PM · Track 4Build the Right Thing: Product Engineering for Software Developers (Part 1)Kent C. DoddsWorkshops Day 112:10 PM · Track 5From Zero to Leaderboard: Building an End-to-End AI Agent Evaluation PipelineWolfram RavenwolfWorkshops Day 112:10 PM · Track 6Local LLMs and workstation agents: Part 2Ahmad OsmanWorkshops Day 112:10 PM · Track 7Beyond RAG: Build a Relational Context Engine from ScratchPeter WerryWorkshops Day 112:10 PM · Track 8Building AI Agents with Real-Time Web DataYohan RajuTrack 812:10 PM · Track 9Research to Reality with Google DeepMindPaige BaileyWorkshops Day 11:15 PM · Track 1Let your agent cook: using skills to evaluate and improve your appAnkur DuggalTrack 11:15 PM · Track 32 hr deep dive on LLM Inference at Scale — Part 2 of 2Harshul Jain, Tanmay SahWorkshops Day 11:15 PM · Track 4Build the Right Thing: Product Engineering for Software Developers — Part 2Kent C. DoddsWorkshops Day 11:15 PM · Track 5Build a Platform, Unleash an Agent on it.... and Watch it Burn!Michael Forrester, Whitney LeeWorkshops Day 11:15 PM · Track 6SonarQube + OpenAI: Wiring Your Team for Agentic DevelopmentKillian Carlsen-PhelanTrack 61:15 PM · Track 7How Reducto parsed the Epstein Files for the Viral JMail Project: The Secret Complexities of DocumentPalak Agarwal1:15 PM · Track 8Turning My Obsidian Vault Into a Local AI EngineerFilip MakraduliWorkshops Day 11:15 PM · Track 9Continuously improving agents with LangfuseLotte Verheyden, Annabell SchäferWorkshops Day 12:20 PM · Track 1From Vibes to Production: Evaluating and Shipping AI Agents That Work 201Laurie VossTrack 12:20 PM · Track 2The Data Context Layer: Why Data Engineering Agents Need More Than Code and DatabasesYoni Michael, Brandon CallenderTrack 22:20 PM · Track 3Special topics in Kernels, RL, Reward Hacking in AgentsDaniel HanWorkshops Day 12:20 PM · Track 4Burn your flags: How PayPal designs interactive CLI tools for agentsMark Lummus, Navinkumar PatilWorkshops Day 12:20 PM · Track 5AI Security Engineer Foundations + CertificateMicah SilvermanWorkshops Day 12:20 PM · Track 6Context Engineering in 2026: Compaction, Memory & CostLouis-François Bouchard, Samridhi Vaid, Omar Solano2:20 PM · Track 7Vector Isn't Enough: Hybrid Search & Retrieval for AI EngineersJeff VestalTrack 72:20 PM · Track 8Build with Perception AgentsEmile Baizel, Shruti AroraWorkshops Day 12:20 PM · Track 9Hands-on AutoResearch: Cracking OpenAI's Parameter GolfZhengyao Jiang, Dixing Xu, Vayum Arora, Dhruv SrikanthWorkshops Day 12:20 PM · Track MObserve, optimize and protect your hosted agents in Microsoft FoundryPamela FoxTrack M4:30 PM · Track 1The Autonomous Computer: Full-stack Infrastructure for Computer Use AgentsAng LiWorkshops Day 14:30 PM · Track 2The Dark Arts of Skill EngineeringPaul Bakaus4:30 PM · Track 4Hill-climbing Skills: How to Improve Agents Without Touching the ModelShubhankar SrivastavaWorkshops Day 14:30 PM · Track 5Agent AuthBereket Habtemeskel, Paola EstefaniaWorkshops Day 14:30 PM · Track 6The Prime Intellect StackWill BrownWorkshops Day 14:30 PM · Track 7Lifestyles of the AI-Native: Voice-coding, agent skills, hooks and scheduled tasksNick Nisi, Zack ProserWorkshops Day 14:30 PM · Track 8The Art and Science of Loopcraft with Pi (and friends)Joel HooksWorkshops Day 14:30 PM · Track 9Evolution of agentic surfacesGagan Bhat, Isabella Kai HeWorkshops Day 15:00 PM · Expo Stage 2 NWHuman Connection in the Age of AIJoyce Zhang, Carole Robin, Ph.D.Expo Stage 26:00 PM · Expo Stage 3 SWExpo Welcome SpeechSonar Sonar, Extend AIExpo Stage 36:15 PM · Expo Stage 3 SWRunway AI Film FestivalExpo Stage 3
Tuesday, June 30
166 sessions9:00 AM · Main StageThe Highest Loopswyx swyxSoftware Factories9:05 AM · Main StageOn AI and KnowledgePablo CastroSoftware Factories9:25 AM · Main StageThe Golden Age of AI EngineeringAlexander Embiricos, Romain HuetSoftware Factories9:45 AM · Main StageGLM-5.2: Frontier Intelligence, Open Weights.Zixuan LiSoftware Factories10:05 AM · Main StageThom Wolf keynoteThom Wolf, Olive SongSoftware Factories10:25 AM · Main StageSecurity Track introManoj NairSoftware Factories10:45 AM · Expo Stage 1 NEEvery AI company is accidentally building a bank.Dor SassonExpo Stage NE10:45 AM · Expo Stage 2 NWThe Enterprise Agentic Gap: When Developer-Level AI Tools Hit Millions of LinesDan Adler10:45 AM · Expo Stage 3 SWHow PayPal Enterprise Payments handles agent-initiated payments across ChatGPT and Google AI ModeSam Parsons10:45 AM · Expo Stage 4 SEAgentic Search for Coding AgentsJakub Hojsan10:45 AM · Leadership 1Inside the AI economy: What Stripe’s data revealsNilofer RajpurkarAgentic Commerce10:45 AM · Leadership 2Governance Is the Real Bottleneck to AI ROIDavid HsuClaws & Personal Agents10:45 AM · Main StageGetting the most out of CodexJason LiuSoftware Factories10:45 AM · Track 1Security Firewall for AgentsRyan DahlClaws & Personal Agents10:45 AM · Track 2The State of VisionJoseph NelsonVision & OCR10:45 AM · Track 3Pinecone 2.0Edo LibertySearch & Retrieval10:45 AM · Track 4Claude Managed Agents Workshop (Part 1)Priyanka Phatak, Gabriel CemajWorkshops Day 210:45 AM · Track 5Through the AI Fog: The architectural decision the next 24 months of agentic security depends on.Manoj NairSecurity10:45 AM · Track 6The New Primitives: Building AI-Native SoftwareKwindla KramerVoice & Realtime AI10:45 AM · Track 7Tokens In, Engagement Out: Training LLM-RecommendersDevansh TandonLLM Recsys10:45 AM · Track 8How Forward Deployed Engineering is done at FactoryEno ReyesForward Deployed Engineering10:45 AM · Track 9Data Quality is the Compute MultiplierAri MorcosData Quality10:45 AM · Track MBuild agents fast with GitHub Copilot (from idea to working app)Idan GazitTrack M11:10 AM · Expo Stage 1 NEGive your coding agents the power of turbogrep!Owen Halpert11:10 AM · Expo Stage 2 NWActionable Knowledge For Agents With Context GraphsWill LyonExpo Stage 211:10 AM · Expo Stage 3 SWFrontier models for the hard parts, open weights for the rest11:10 AM · Expo Stage 4 SEAgents, codebases, and teams: what it actually takes to ship togetherAditya Khandelwal11:10 AM · Leadership 1Building the engine while flying the plane — launching the Figma MCP serverJesse LumarieAI-Native Enterprises11:10 AM · Leadership 2Your Agent Evolved. Your Evals Didn't.Ameya BhatawdekarAI Architects: Show my Workflow11:10 AM · Main StageRise of the Software FactoryTereza TížkováSoftware Factories11:10 AM · Track 1Your Agent Didn’t Fail. Your Harness Did.Vinoth GovindarajanClaws & Personal Agents11:10 AM · Track 2Building the Document Context Layer for AI AgentsJerry LiuVision & OCR11:10 AM · Track 3The unreasonable effectiveness of BM25 for agentic searchJo Kristian BergumSearch & Retrieval11:10 AM · Track 4Claude Managed Agents workshop (Part 2)Priyanka Phatak, Gabriel CemajWorkshops Day 211:10 AM · Track 5Your LLM Stack Is a 2008 Database With Better Marketing: Why ML Security Is Dominated by Misconfiguration, Not Missing FeaturesLovina DmelloSecurity11:10 AM · Track 6Speech-to-Speech Model Research at Google DeepMindValeria Wu Fon, Tom OuyangVoice & Realtime AI11:10 AM · Track 7Spotify LLM RecsysJacqueline Wood, Yves RaimondLLM Recsys11:10 AM · Track 8How Forward Deployed Engineering is done at CursorPauline BrunetForward Deployed Engineering11:10 AM · Track 9The Messy Reality of Scale: Synthetic Data and Pre-Training at PoolsideRobert McHardy, Marah AbdinData Quality11:40 AM · Expo Stage 2 NWAgentic vs. Vector Search: An Eval-Driven Approach to Coding Agent PerformanceJess Wang11:40 AM · Expo Stage 3 SWAgents Don't Have Coworkers, They Have HostagesGabriel Martinez11:40 AM · Expo Stage 4 SEWould your AI agent get the job? A performance review framework for enterprise agentsAndreea Pleşea, Dan Bălăceanu11:40 AM · Leadership 1Agentic SDLC at Uber: Building Blocks for Uber's Software FactoryUday Kiran Medisetty, Adam HudaAI-Native Enterprises11:40 AM · Leadership 2The Last Human Code Review: Building Trust in AI-Generated CodeItamar FriedmanAI Architects: Show my Workflow11:40 AM · Main StageOrchestras, not FactoriesCharlie HoltzSoftware Factories11:40 AM · Track 1Everyone Gets A Software CompanyBenjamin Guo, Rob CheungClaws & Personal Agents11:40 AM · Track 2Skill issue: stop deploying vision language models, use them with Skills to build e2e vision apps on edgeMerve NoyanVision & OCR11:40 AM · Track 3The Search Engine for the Agentic WebWill BrykSearch & Retrieval11:40 AM · Track 4Claude Managed Agents workshop (Part 3)Priyanka Phatak, Gabriel CemajWorkshops Day 211:40 AM · Track 5We Gave an Agent Production Code Access and Then Tried to Sleep at NightMoritz JohnerSecurity11:40 AM · Track 6Voice Agents Can Just Do ThingsCharlie GuoVoice & Realtime AI11:40 AM · Track 7LLM Recsys at DoorDashRaghav SabooLLM Recsys11:40 AM · Track 8AI tools for Forward Deployed EngineeringVasuman MozaForward Deployed Engineering11:40 AM · Track 9Rethinking Environments for Long Horizon WorkRayan GargData Quality11:40 AM · Track MUse Copilot across CLI, dev, and cloud workflows to move faster end-to-endPamela FoxTrack M12:05 PM · Expo Stage 1 NEYour Agent Is Lying to You About Whether It WorkedDat Ngo12:05 PM · Expo Stage 2 NWWhy building building agent quality platforms is hard.Hossein Niazmandi12:05 PM · Expo Stage 3 SWCan LLMs write fast multi-GPU kernels? We built a benchmark to find out.Simran Arora12:05 PM · Expo Stage 4 SESelf-Improving Agents That Teach the Company BackRafal Wilinski12:05 PM · Leadership 1Scaling Code Quality: Building uReview, Uber’s Multi-Agent Code Review EngineWill Bond, Ameya KetkarAI-Native Enterprises12:05 PM · Leadership 2Prototyping as Leadership: How a CTO Ships with AI AgentsHursh AgrawalAI Architects: Show my Workflow12:05 PM · Main StageWhat we learned by analyzing 1M AI-generated PRsDaksh GuptaSoftware Factories12:05 PM · Track 1Tethered: Our Agents Are UsShu FangClaws & Personal Agents12:05 PM · Track 2Modality Misalignment and Originality Attribution in Short-Form Video: A Multi-Agent Approach at Platform ScaleAditya GautamVision & OCR12:05 PM · Track 3Rebuilding the web for agentsLiad YosefSearch & Retrieval12:05 PM · Track 4Claude Managed Agents workshop (Part 4)Priyanka Phatak, Gabriel CemajWorkshops Day 212:05 PM · Track 5Agentic Development SecurityEzra TanzerSecurity12:05 PM · Track 6Your Voice Agent is Just a Walkie-TalkieNeil ZeghidourClaws & Personal Agents12:05 PM · Track 7Open Q&A: LLM RecsysDevansh TandonLLM Recsys12:05 PM · Track 8How Forward Deployed Engineering is done at CognitionJia WuForward Deployed Engineering12:05 PM · Track 9Bugcrowd posttraining talkDavid BrumleyPosttraining & Midtraining1:30 PM · Expo Stage 1 NEEvery Agent, Everywhere, All at OnceVlad Luzin1:30 PM · Expo Stage 3 SWDesigning Evals That Earn User TrustFelipe Blanes1:30 PM · Expo Stage 4 SEStop promptingGreg PstruchaExpo Stage 41:30 PM · Leadership 1Spin at the Gate Until Green: The Engineering Primitives Behind Self-Driving CodebasesAndrew OrobatorSoftware Factories1:30 PM · Leadership 2Serving 2 Million Models Without Melting: Scaling the Hugging Face HubArek BoruckiAI Architects: Show my Workflow1:30 PM · Main StageGet Out of the Model's WayKevin HouSoftware Factories1:30 PM · Track 1Agents' next frontier: agent-to-agent and network effectsJean-Denis GrezeClaws & Personal Agents1:30 PM · Track 2From Ingestion to Agents: How Leading AI Teams Build on Document IntelligenceAdit AbrahamVision & OCR1:30 PM · Track 3If we want them to do Knowledge Work, we need to design Knowledge AgentsBenjamin ClaviéSearch & Retrieval1:30 PM · Track 4Everybody Gets a Digital Clone! (Part 1 of 3)Neil ZeghidourWorkshops Day 21:30 PM · Track 5Using LLMs to Secure Source CodeEugene YanSecurity1:30 PM · Track 6Tolan: Voice-First AI CompanionPaula DozsaVoice & Realtime AI1:30 PM · Track 7From approval loops to autonomous agents with Docker pt1John CraftLLM Recsys1:30 PM · Track 8The Dirty Secret of Forward Deployed EngineeringNatalie MeurerForward Deployed Engineering1:30 PM · Track 9The Base Model is DeadVarun SinghData Quality1:30 PM · Track MModernize CI/CD using agent-assisted workflows that reduce manual debuggingSalil SubbakrishnaTrack M1:55 PM · Expo Stage 1 NEWho Approved That MCP Server? Governing the Tool LayerJim Clark1:55 PM · Expo Stage 2 NWVoice Agents Are Mostly Invisible. Here's How to See Them.Fuad Ali1:55 PM · Expo Stage 3 SWwhat we learned by analyzing 1M AI generated PRs1:55 PM · Expo Stage 4 SEDeploying browser agents at scaleDerek Meegan1:55 PM · Leadership 1AI Evals Platform for Cross-Functional Teams at ScaleNachiket Paranjape, Swaroop Chitlur HaridasAI-Native Enterprises1:55 PM · Leadership 2IT Admin for the AI Workforce: Why Your AI Agents Will Need Their Own IT DepartmentSarthak AggarwalAI Architects: Show my Workflow1:55 PM · Main StageSelf-Improving software factories: The new open source model"Zach LloydSoftware Factories1:55 PM · Track 1Claude for long-horizon tasksLance MartinClaws & Personal Agents1:55 PM · Track 2The Best Models Still Reason Like ToddlersAndrew DaiVision & OCR1:55 PM · Track 3Your Agreements Are a Database You Can't Query. We're Fixing ThatHiral Shah, Sean SodhaSearch & Retrieval1:55 PM · Track 4Everybody Gets a Digital Clone! (Part 2 of 3)Neil ZeghidourWorkshops Day 21:55 PM · Track 5Dual-Surface Architecture: Serving Humans and Agents from the Same Tool LayerEthan (Jung Min) ChaSecurity1:55 PM · Track 65 Voice Agent Failure Modes You'll Hit in Week OneVenky B, Vyas AVoice & Realtime AI1:55 PM · Track 7From approval loops to autonomous agents with Docker pt2John CraftLLM Recsys1:55 PM · Track 8How Forward Deployed Engineering is done at DecagonSunny RekhiForward Deployed Engineering1:55 PM · Track 9Ending AI SlopThais Castello BrancoData Quality2:25 PM · Expo Stage 1 NEBeyond Golden Signals: Monitoring in the Age of GenAIMarina Petzel2:25 PM · Expo Stage 2 NWBuild agents fast with GitHub Copilot (from idea to working app)Idan GazitExpo Stage 22:25 PM · Expo Stage 3 SWBuilding agents is trivial now, context is the next frontierJeff Ng2:25 PM · Expo Stage 4 SEContinuous Engineering: Software Development for the Age of Agents2:25 PM · Leadership 1Productionizing LLM Gateways: Architecture, Tradeoffs, and Hard Lessons from the TrenchesKanish ManujaAI-Native Enterprises2:25 PM · Leadership 2The Era of Compound EngineeringKieran KlaassenAI Architects: Show my Workflow2:25 PM · Main StageWe're the bottleneck, but we don't have to beIdo SalomonSoftware Factories2:25 PM · Track 1From coding to Knowledge work agentsKaran VaidyaClaws & Personal Agents2:25 PM · Track 2You’re Not Thinking Big Enough: Rebuilding Food Systems from First Principles with AI AgentsCody MenefeeVision & OCR2:25 PM · Track 3How to Connect AI to Billions of Legal DocumentsSimon Eskildsen, Jacob LauritzenSearch & Retrieval2:25 PM · Track 4Everybody Gets a Digital Clone! (Part 3 of 3)Neil ZeghidourWorkshops Day 22:25 PM · Track 5Agentic Security: Permissions, Provenance, and the Agent Supply ChainSteve YeggeSecurity2:25 PM · Track 6I Monitored Crime Audio. Voice Agents Scare Me More.Sumanyu SharmaVoice & Realtime AI2:25 PM · Track 7From approval loops to autonomous agents with Docker pt3John CraftLLM Recsys2:25 PM · Track 8How Forward Deployed Engineering is done at RampLeo MehrForward Deployed Engineering2:25 PM · Track 9Scaling to Long-Horizons: Algorithms, Environments, ComputeRoss Taylor, Chengxi TaylorData Quality2:25 PM · Track MUsing AI tools to teach old apps new tricksMaria BledsoeTrack M2:50 PM · Expo Stage 1 NE6 Pillars of an Agentic Harness That Fixes Production IncidentsVarun Krovvidi2:50 PM · Expo Stage 2 NWVideo Discovery for Agentic World-Model TrainingRafael Levi2:50 PM · Expo Stage 4 SESelf-Driving Production: AI Wrote your Code. AI Should Fix It, Too2:50 PM · Leadership 1From AI-Assisted to AI-Native: Building a Frontier Development TeamClare LiguoriAI-Native Enterprises2:50 PM · Leadership 2How I automate my own job at Hugging Face using agentsNiels RoggeAI Architects: Show my Workflow2:50 PM · Main StageNotion's Token TownSarah SachsSoftware Factories2:50 PM · Track 1Your company brain will leak secrets. Here's how we stopped it for big banks and ourselves.Tanmai GopalClaws & Personal Agents2:50 PM · Track 2From VLM/VLA's to Embodied AgentsArmen AghajanyanVision & OCR2:50 PM · Track 3Where RL Will Take SearchMaximilian-David Rumpf, Lotte SeifertSearch & Retrieval2:50 PM · Track 4Setting Yourself Up for Success — Part 1Jason LiuWorkshops Day 22:50 PM · Track 5It's 10pm. Do You Know Where Your Agents Are?Kim MaidaSecurity2:50 PM · Track 6Realtime Voice Agents with Frontier IntelligenceBohan LiVoice & Realtime AI2:50 PM · Track 7From approval loops to autonomous agents with Docker pt4John CraftLLM Recsys2:50 PM · Track 8Forward Deployed Engineering 101Kevin BaiForward Deployed Engineering2:50 PM · Track 9When Will The Benchmaxxing Plague End?Nick HeinerAI Architects: Show my Workflow3:20 PM · Expo Stage 1 NECan Your Agent Hear You Now?Thor 雷神 SchaeffExpo Stage 33:20 PM · Expo Stage 2 NWFrom Context to Memory: Your Agents Need a Real Memory LayerAnders Swanson3:20 PM · Expo Stage 3 SWRunning a 20T-Token Data Pipeline: Infrastructure Lessons from ProductionBogdan Gaza3:20 PM · Expo Stage 4 SEFrom raw documents to AI-ready dataLeo Platzer3:20 PM · Leadership 1How to Get Your Org to Adopt Coding Agents (Without Shipping Garbage)Eyal BlumAI-Native Enterprises3:20 PM · Leadership 2Your Fine-Tuned Model Is Tech Debt: A 50x ROI House of CardsDan BjornnAI Architects: Show my Workflow3:20 PM · Main Stagefighting slop with slopVaibhav GuptaSoftware Factories3:20 PM · Track 1Every Harness Will Become A ClawSam BhagwatClaws & Personal Agents3:20 PM · Track 2From Scratch to SOTA: Training a 3B State-Space Vision Model for 1.4 Billion PeopleKrishna Prasad SrinivasanVision & OCR3:20 PM · Track 3Stop Chunking Like It's 2022Yuval Belfer, Niv GranotSearch & Retrieval3:20 PM · Track 4Setting Yourself Up for Success — Part 2Jason LiuWorkshops Day 23:20 PM · Track 5AI’s Jurassic Park PeriodAaron StanleySecurity3:20 PM · Track 6"My name is... my name is...": A Linguistic Map for Building and Debugging Voice AgentsMidam KimVoice & Realtime AI3:20 PM · Track 7From approval loops to autonomous agents with Docker pt5John CraftLLM Recsys3:20 PM · Track 8How Forward Deployed Engineering is done at KeplerVinoo GaneshForward Deployed Engineering3:20 PM · Track 9Building Worlds for ModelsNicolai OuporovData Quality3:20 PM · Track MSurviving Your Own Velocity: How VS Code Ships Weekly with 40 PeopleHarald KirschnerTrack M3:45 PM · Expo Stage 1 NEHow We Built the Airbyte Agent MCP Server and CLIPedro LopezExpo Stage 13:45 PM · Expo Stage 2 NWFrom Chatbots to Agents: How Reducto builds for Agent Experience to Enable Real WorkAbhi Arya3:45 PM · Expo Stage 3 SWTowards Reliable Financial Agents: How a 4B Model Outsmarted a 235B GiantCharlie Dickens3:45 PM · Expo Stage 4 SEAI Enablement at Automattic: How a Remote Company Builds AI FluencyEm Shreve3:45 PM · Leadership 2Unlock Agent Autonomy: The Runtime for AI-Native SystemsTushar JainAI Architects: Show my Workflow3:45 PM · Main StageLoop Engineering from first principlesKyle MisteleSoftware Factories3:45 PM · Track 1Gadgets: Personal app vibe coding that is actually safeKenton VardaSoftware Factories3:45 PM · Track 4Setting Yourself Up for Success — Part 3Jason LiuWorkshops Day 23:45 PM · Track 5Secure Cloud ComputeEthan SutinSecurity3:45 PM · Track 6Act, Confirm, or Stop? Smarter behavior for AI assistants, wearables & robotsAmit DesaiVoice & Realtime AI3:45 PM · Track 9Data and Environment Curation for Post-training LLMsMahesh SathiamoorthyData Quality4:30 PM · Main StageHarness Engineering is not Enough: Why Software Factories FailDex HorthySoftware Factories4:50 PM · Main StageIn Code They Act, In Proof We TrustErik MeijerHarness Engineering5:10 PM · Main StageRecursive Model ImprovementLee RobinsonSoftware Factories
Wednesday, July 1
166 sessions9:05 AM · Main StageField Guide to FableThariq ShihiparAutoresearch9:25 AM · Main StageIn the Land of AI Agents, the Verifiers Are KingTariq ShaukatSoftware Factories9:45 AM · Main StagePerception AgentsAntje BarthAutoresearch10:05 AM · Main StageResearch to Reality with Google DeepMindBenoit SchillingsAutoresearch10:25 AM · Main StageEvals Track IntroLaurie Voss, Aparna DhinakaranAutoresearch10:45 AM · Expo Stage 2 NWAI Engineering & Governance 2026 TrendsWallon WalusayiExpo Stage 210:45 AM · Expo Stage 3 SWWhy AI Didn't Actually Make You Ship FasterGabriel Spencer-HarperExpo Stage 310:45 AM · Expo Stage 4 SERedesigning how software gets builtTBD — SonarExpo Stage 410:45 AM · Leadership 1How do you diffuse AI into the real world?Varun ShenoyAI-Native Enterprises10:45 AM · Leadership 2The Z/L Continuum: Should AI Engineers Still Read Code?Alex VolkovAI Architects: Tokenmaxxing10:45 AM · Main StageFirst Steps Toward Automated AI ResearchRichard SocherAutoresearch10:45 AM · Track 1Don’t build agents, build environmentsAdam AzzamSandbox & Platform Engineering10:45 AM · Track 2Building the simulation infrastructure for practical world model useChristopher ManningRobotics & World Models10:45 AM · Track 3Beyond Static Intelligence: Evaluating Continual LearningParth AsawaMemory & Continual Learning10:45 AM · Track 4Build realtime multimodal agents with Gemini LiveThor 雷神 SchaeffWorkshops Day 210:45 AM · Track 5Vending-Bench: Long-Horizon Agent Evals for a Simulated Vending BusinessLukas PeterssonEvals10:45 AM · Track 6Understanding is the new bottleneckGeoffrey LittDesign Engineering10:45 AM · Track 7Computer-use models will agentify the web, not APIsDhruv BatraComputer Use10:45 AM · Track 8Build-Time vs. Run-Time: Why Your Dev Tools Will Fail in ProductionAveri Kitsch, Prerna KakkarContext Engineering10:45 AM · Track 9What's next after RLHF?Diogo AlmeidaPosttraining & Midtraining10:45 AM · Track MFrom framework to runtime: running agents with Foundry Agent ServiceTina Manghnani, Keiji KanazawaTrack M11:00 AM · Leadership LoungeTokenomics: From AI Spend to AI ValueMartin Harrysson, Matt Linderman, Prakhar DixitCTO Circle11:10 AM · Expo Stage 1 NEHarnessing Agents: The Durable Runtime for Dynamic WorkflowsViren Baraiya11:10 AM · Expo Stage 2 NWAI-Assisted Engineering: 5 Trends We're Seeing From 500+ OrganizationsJustin ReockExpo Stage 211:10 AM · Expo Stage 3 SWThe Death of Keyword Search and the Rise of Agent-Readable CatalogsNixon DinhExpo Stage 311:10 AM · Expo Stage 4 SEFDE Playbook: Build an AI Support Agent and Give It a VoiceMatt LawlerExpo Stage 411:10 AM · Leadership 1How to avoid disaster when vibe-coding a billing engineAndrew GarvinAI-Native Enterprises11:10 AM · Leadership 2Is Orchestration the Future?Vlad LuzinAI Architects: Tokenmaxxing11:10 AM · Main StageAutoresearch for Dense Retrieval: Test-Time Compute with Frozen Embedding ModelsHan XiaoAutoresearch11:10 AM · Track 1Letting the Interns Loose — How We Accelerated AI Adoption.Shashank GoyalSandbox & Platform Engineering11:10 AM · Track 2Building the simulation infrastructure for practical world model use (Part 2)Christopher ManningRobotics & World Models11:10 AM · Track 3Scaling up Continual LearningRonak MaldeMemory & Continual Learning11:10 AM · Track 4Build realtime multimodal agents with Gemini Live (continued 2)Thor 雷神 SchaeffWorkshops Day 211:10 AM · Track 5From Signal to PR: Anatomy of a Self-Improving AgentJason LopateckiEvals11:10 AM · Track 6The Spatial Harness: Bringing Agents to the CanvasMax DrakeDesign Engineering11:10 AM · Track 7Computer Use at the Edge of the Statistical PrecipicePierluca D'OroComputer Use11:10 AM · Track 8It’s Tokens All The Way Down: How RLMs are DifferentKevin MaduraContext Engineering11:10 AM · Track 9State of DataSean CaiPosttraining & Midtraining11:40 AM · Expo Stage 1 NEFault-Tolerant Training at Scale: Making Hardware Failures a Non-Event11:40 AM · Expo Stage 2 NWHow to generate mergeable code with a context enginePeter Werry11:40 AM · Expo Stage 4 SEAI agents don't read your policy docs. They hit your APIs.11:40 AM · Leadership 1Your Code Has Bugs. Lean4 Has Proofs. A Practical Guide to Formal Verification for EngineersVarun PantAI-Native Enterprises11:40 AM · Leadership 2How to Kill the Code ReviewAnkit JainAI Architects: Tokenmaxxing11:40 AM · Main StageMemory Harnesses for Long-Running Research AgentsStefania DrugaMemory & Continual Learning11:40 AM · Track 1Kubernetes Is Not Your SandboxIvan BurazinSandbox & Platform Engineering11:40 AM · Track 2Commercial Grade-Robots for Real World UsageJason MaRobotics & World Models11:40 AM · Track 3Scaling Compute on ContextJack MorrisMemory & Continual Learning11:40 AM · Track 4Build realtime multimodal agents with Gemini Live (continued 3)Thor 雷神 SchaeffWorkshops Day 211:40 AM · Track 5Building Closed-Loop Evals for a Multimodal Agent at Uber ScaleSoumya Gupta, Jai ChopraEvals11:40 AM · Track 6The Design-Code Roundtrip That Isn'tJonathan GordonDesign Engineering11:40 AM · Track 7Bringing agents onto the world wide webPaul Klein IVComputer Use11:40 AM · Track 8500 Skills, Zero Fine-Tuning: LinkedIn's Playbook for AI Agents That Actually Know Your CodebaseAjay PrakashContext Engineering11:40 AM · Track 9Training Frontier Models to Out-Think HackersUri Rolls, Thom WolfData Quality11:40 AM · Track MOpenAI, Anthropic, or agent frameworks: choose the right AI stackArun Sekhar, Pamela FoxTrack M12:05 PM · Expo Stage 1 NEYour agent architecture has a half-life of 6 monthsDan Farrelly12:05 PM · Expo Stage 2 NWFrom Stateless to Stateful: Orchestrating Real-Time Voice & Messaging Agents with Twilio and Amazon BedrockRishab Kumar12:05 PM · Expo Stage 3 SWHarnessing Collective Agent Intelligence for Open ScienceJames Zou12:05 PM · Expo Stage 4 SEPrompt, Memory, Weights: The Architecture Decisions Most AI Teams Make by AccidentAnant SrivastavaContext Engineering12:05 PM · Leadership 1AI-Native Organisations runs on Skills: How to Extract, Structure, evaluate and Scale ThemImad TouilAI-Native Enterprises12:05 PM · Leadership 2The Death of the Code ReviewLaurie VossAI Architects: Tokenmaxxing12:05 PM · Main Stage« the era of (auto) research »Elie BakouchAutoresearch12:05 PM · Track 1Your agent needs a sandbox, not a desertSamuel ColvinSandbox & Platform Engineering12:05 PM · Track 3Intelligence + Continual Learning = ExpertiseYu SuMemory & Continual Learning12:05 PM · Track 4Build realtime multimodal agents with Gemini Live (continued 4)Thor 雷神 SchaeffWorkshops Day 312:05 PM · Track 5From Agent Traces to Agent Simulations: The next era of agent evaluationRustem FeyzkhanovEvals12:05 PM · Track 6Mousepower: agents that can’t be measured, can’t be managed.Maximillian PirasDesign Engineering12:05 PM · Track 7The Dark Arts of Web Automation: Teaching Agents to Use Websites Like HumansCorey GallonComputer Use12:05 PM · Track 8Your agents lack context: Here's how to fix "You're absolutely right!"Brandon WaselnukContext Engineering12:05 PM · Track 9Learning on the job: the future of post-trainingRaymond FengPosttraining & Midtraining1:30 PM · Expo Stage 2 NWSurviving Your Own Velocity: How VS Code Ships Weekly with 40 PeopleHarald KirschnerExpo Stage 21:30 PM · Expo Stage 3 SWWhy Agents Should Have Their Own SandboxPhilipp SchmidExpo Stage 11:30 PM · Leadership 1The Half Life of Agent InfrastructureBen KusAI-Native Enterprises1:30 PM · Leadership 2Tokenmaxxing is the New "Lines of Code"Nicholas ArcolanoAI Architects: Tokenmaxxing1:30 PM · Main StageClosing the Loop: An Autonomous AI Research AgentTim SweeneyAutoresearch1:30 PM · Track 1From fork() to Fleet: Designing an Agent Sandbox Cloud Pt 1Abhishek BhardwajSandbox & Platform Engineering1:30 PM · Track 2Unitree: Building Mass Produced HumanoidsXiangMing SunRobotics & World Models1:30 PM · Track 3Adaption Labs — Gradient-Free Continual LearningSara HookerMemory & Continual Learning1:30 PM · Track 4The Agentic Power User's Playbook: Tips and Tricks for Swarm-Style Agentic DevelopmentJohn LindquistWorkshops Day 31:30 PM · Track 5Model Whisperers: How Evals and Prompts Shape Agent BehaviorChris Souza, Preetika Bhateja, Daniel BumpEvals1:30 PM · Track 6Design at the Speed of AdjectivesPaul BakausDesign Engineering1:30 PM · Track 7From RL to IRLGaurav MishraComputer Use1:30 PM · Track 8How long can your skills be before your agent forgets what you told it?Laurie VossContext Engineering1:30 PM · Track 9Reinforcement Learning without Verifiable RewardsWill BrownPosttraining & Midtraining1:55 PM · Expo Stage 1 NEEdge-Native AI: Building Ultra-Fast Agents and MCP Servers with SpinThorsten HansExpo Stage 21:55 PM · Expo Stage 2 NWWhy your company needs a context graph, and how to build itGil FeigExpo Stage 31:55 PM · Expo Stage 3 SWWarp: Building Self-Improving Agent Software FactoriesSuraj Gupta1:55 PM · Expo Stage 4 SENatively Multimodal from Step Zero1:55 PM · Leadership 1Guardians of the State: How We Built an Air-Gapped AI Fortress for Consumer DataRachna SrivastavaAI-Native Enterprises1:55 PM · Leadership 2Engineering Agency out of the Happy PathMatthew JewkesAI Architects: Tokenmaxxing1:55 PM · Main StageAn AI Agent Became the #1 Contributor in OpenAI's Hiring ChallengeZhengyao JiangAutoresearch1:55 PM · Track 1From fork() to Fleet: Designing an Agent Sandbox Cloud Pt2Abhishek BhardwajSandbox & Platform Engineering1:55 PM · Track 2Frontier Robotics ResearchDeepak PathakRobotics & World Models1:55 PM · Track 3Improving Agents is a Data Mining ProblemVivek TrivedyMemory & Continual Learning1:55 PM · Track 4The Agentic Power User's Playbook: Tips and Tricks for Swarm-Style Agentic Development (continued 2)John LindquistWorkshops Day 31:55 PM · Track 5Evaling Video SlopMaor BrilEvals1:55 PM · Track 6Training TasteThais Castello BrancoDesign Engineering1:55 PM · Track 7The Rise of CaaS: Context-as-a-Service for Agentic AIOmer PrimorComputer Use1:55 PM · Track 8WTF Is the Context Layer? The Missing Infrastructure for Production AgentsPrukalpa SankarContext Engineering1:55 PM · Track 9Emulated: The data for fully autonomous software engineers and companiesJoseph WangPosttraining & Midtraining2:25 PM · Expo Stage 1 NEPower agents with Microsoft IQRonak ChokshiExpo Stage 12:25 PM · Expo Stage 2 NWBeyond Code Generation: API Context for Agentic EngineeringKamalakannan Nandagopal2:25 PM · Expo Stage 3 SWLatency Is a Budget. Humanlike Is the Goal.Jesse Hall2:25 PM · Expo Stage 4 SEYour Stack Has a Latency Problem You Can’t See2:25 PM · Leadership 1From Tokenmaxxing to Trusted ThroughputMingsheng HongAI-Native Enterprises2:25 PM · Leadership 2I Let Agents Refactor My Codebase for 3 Weeks. Then I Read the Code.Keiji KanazawaAI Architects: Tokenmaxxing2:25 PM · Main StageSelf-Improvement of Context, Harness, and Model Weights through Reflective OptimizationLakshya AgrawalAutoresearch2:25 PM · Track 11,000 Agent Tasks in a Sandbox: What Breaks When LLMs Write and Run CodeKevin OrellanaSandbox & Platform Engineering2:25 PM · Track 2From Manual Drones to Autonomous Multi-Agent MissionsSuchet BargotiRobotics & World Models2:25 PM · Track 3Bringing Continual Learning into EnterprisesSamuel DentonMemory & Continual Learning2:25 PM · Track 4The Agentic Power User's Playbook: Tips and Tricks for Swarm-Style Agentic Development (continued 3)John LindquistWorkshops Day 32:25 PM · Track 5Ask YouTube — Open Q&AMihnea MunteanuEvals2:25 PM · Track 6Imagination EngineeringEve BouffardDesign Engineering2:25 PM · Track 7Computer-Use 2.0: Agents Just Got Multi-CursorFrancesco Bonacci, Dillon DuPontComputer Use2:25 PM · Track 8MCP Apps - Extending the frontierLiad Yosef, Ido SalomonContext Engineering2:25 PM · Track 9LatchBioKenny WorkmanPosttraining & Midtraining2:25 PM · Track MPower agents with Microsoft IQMarco CasalainaTrack M2:50 PM · Expo Stage 1 NEInference performance as a competitive advantageAlex Campos, Yunmo Koo2:50 PM · Expo Stage 2 NWBuilding an Agent Harness for the Business, Not the BuilderGarrett Galow2:50 PM · Expo Stage 3 SWThe Frontier Is Coming HomeDylan Couzon2:50 PM · Expo Stage 4 SEContinuous Offensive Security the only approach in an agent-first worldEli Cohen2:50 PM · Leadership 1Agents Are Where Microservices Were in 2015. We're Making All the Same Mistakes.Roberto Milev, Uday KanagalaAI-Native Enterprises2:50 PM · Leadership 2Intelligent Model Routing: Frontier Performance Without Frontier BillsTomás Hernando KofmanSandbox & Platform Engineering2:50 PM · Main StageAutoresearch for KernelsTejas BhaktaAutoresearch2:50 PM · Track 1The Next Trillion Users of the Internet Still Don't Have an IdentityAdi SinghSandbox & Platform Engineering2:50 PM · Track 2Why Large? Tiny LMs & Agents on Edge/RoboticsCormac BrickRobotics & World Models2:50 PM · Track 3Designing Agents (The Floor Is the Frontier)Ben HylakMemory & Continual Learning2:50 PM · Track 4Don't Write Skills, Train ModelsBrian Douglas, John McBrideWorkshops Day 32:50 PM · Track 5Evals Driven-Development: Engineering a Mental Health AI Coach Ethically & SafelyAkele Reed, Dave Revere, Doug KellerEvals2:50 PM · Track 6The Missing Layer: Design Taste in AI Agents // Stop Letting Your Agents Ship Ugly UIsHassan El MghariDesign Engineering2:50 PM · Track 7Will AI predict people like we predict the weather? (alternate title “A field guide to synthetic personas for market research”)Ishan AnandComputer Use2:50 PM · Track 8MCP Apps: Give the Model Data, Give the User a UIDustin MihalikContext Engineering2:50 PM · Track 9Agents at Scale: Inside MiniMax's Model and the Infrastructure Behind ItOlive Song, Dan FuPosttraining & Midtraining3:20 PM · Expo Stage 1 NEThe Self-Improving OSS Agent Stack3:20 PM · Expo Stage 2 NWAI Applications in a flash! No Dev Ops. Just code.Dean Quiñanola3:20 PM · Expo Stage 3 SWThe Infinite Context Window Is a Myth: Context Engineering for AI AgentsElizabeth Fuentes Leone3:20 PM · Expo Stage 4 SEVibe Code Safely: Introducing Gadgets3:20 PM · Leadership 1Agentic Sites: Building Hyper Personalized WebsitesCarlos SanchezAI-Native Enterprises3:20 PM · Leadership 2Inference is the New Training Loop: Architecting High-Reliability Agents and Continuous AI SystemsDavid CorbittPosttraining & Midtraining3:20 PM · Main StageAutoresearch in the wildRoland Gavrilescu, Julian BrightAutoresearch3:20 PM · Track 1Sandboxes Aren't Optional: Runtime Isolation Patterns for Coding Agents at ScaleRobert BrennanSandbox & Platform Engineering3:20 PM · Track 2From Self-Driving Monorepo to Self-Driving CarsAmit NavindgiRobotics & World Models3:20 PM · Track 3Lessons from Studying Every Memory SystemShlok KhemaniMemory & Continual Learning3:20 PM · Track 4Don't Write Skills, Train Models (cont. 2/3)Brian DouglasWorkshops Day 33:20 PM · Track 5Don't Ship Skills Without EvalsPhilipp SchmidEvals3:20 PM · Track 6Generative UI... in Python?Jeremiah LowinDesign Engineering3:20 PM · Track 7How Web Data Infrastructure Powers the Next Generation of AIPatricija ŽemaitytėComputer Use3:20 PM · Track 8MCP Tasks (async)/ Why the heck aren't any agents supporting MCP tasks/async?Cornelia DavisContext Engineering3:20 PM · Track 9Benchmarks: The Good, the Bad, and the UglyAli KhialPosttraining & Midtraining3:20 PM · Track MDeploy agents to users in M365, Teams, and appsAshu JoshiTrack M3:45 PM · Expo Stage 1 NEModular: Taming the AI Hardware Cambrian ExplosionAbdul Dakkak3:45 PM · Expo Stage 2 NWBuilding on the Codex HarnessDominik Kundel3:45 PM · Expo Stage 3 SWStop Renting Intelligence: The Train-to-Deploy Loop for Specialized AIJetashree Ravi3:45 PM · Expo Stage 4 SERay Actors, Vision Tokens, and the GIL: Engineering an SFT Data Pipeline That Keeps GPUs BusyTarun Sunkaraneni3:45 PM · Leadership 1The Chief AI Officer: A framework for the emerging Swiss Army Knife of rolesRania KhalafAI Architects: Tokenmaxxing3:45 PM · Leadership 2The state of AI in software development: Insights across 400+ organizationsJustin ReockAI Architects: Tokenmaxxing3:45 PM · Main StageAutoresearch in a Multi-Agent AI VillageErina Karati, Arunachalam ManikandanAutoresearch3:45 PM · Track 1Building ambitious softwareJonathan KelleySandbox & Platform Engineering3:45 PM · Track 2I gave an AI a bodyCyrus ClarkeRobotics & World Models3:45 PM · Track 3LLM Knowledge Bases: a practical guideBen HolmesMemory & Continual Learning3:45 PM · Track 4Don't Write Skills, Train Models (cont. 3/3)Brian DouglasWorkshops Day 33:45 PM · Track 5Everything Is a RolloutAlex Shaw, Ryan MartenEvals3:45 PM · Track 6One Designer + Al. Hundreds of Deliverables.Vincent WendyDesign Engineering3:45 PM · Track 8The Universal Remote Control for AIAlex HancockContext Engineering4:30 PM · Main StageClosing KeynoteAddy OsmaniAutoresearch4:50 PM · Main StageTrends in AIGeorge Cameron, Micah Hill-SmithAutoresearch5:10 PM · Main StageClosing KeynoteWei-Lin ChiangAutoresearch
Thursday, July 2
171 sessions8:30 AM · Networking RoomBattlefield setupStartup Battlefield9:00 AM · Main StageThe 2026 State of AI EngineeringBarr YaronHarness Engineering9:20 AM · Main StageTCP and RDMA are Killing Inference Throughput; Homa can Fix ItJohn OusterhoutSoftware Factories9:40 AM · Main StageThe Unreasonable Effectiveness of Separating the Task from the ModelMaxime Rivest, Isaac MillerHarness Engineering10:00 AM · Main StageHow Anthropic Builds: Lessons from LabsMike Krieger, swyx swyxHarness Engineering10:20 AM · Main StageThinner Agents on a Smarter Substrate: The Ontology-based Semantic LayerEmil EifremGraphs10:45 AM · Expo Stage 1 NEAI Engineering & Governance 2026 TrendsWallon WalusayiExpo Stage 110:45 AM · Expo Stage 2 NWYour Agent Can't Tell If It's RightWillem Pienaar10:45 AM · Expo Stage 3 SWNo, That's Not a Software FactoryRyan CookeExpo Stage 310:45 AM · Expo Stage 4 SEVector Isn't Enough: Hybrid Search & Retrieval for AI Engineers10:45 AM · Leadership 1Building safe payment infrastructure for machine-to-machine commerceJennifer LeeAgentic Commerce10:45 AM · Leadership 2The Agent Behind the Curtain: Building the Oz Cloud Agent PlatformSafia AbdallaAI Architects: AI Factories10:45 AM · Main StageTokens Should Have JobsKatelyn Lesse, Angela JiangHarness Engineering10:45 AM · Networking Room$100,000 AIE Startup Battlefield — presented by HyperagentHowie LiuStartup Battlefield10:45 AM · Track 1Training Krea 2 - What matters in generative model training.Sangwu LeeGenerative Media10:45 AM · Track 2Designing Multimodal Collaborative Agents for Next-Gen CommerceNidhi Kaushik VyasAgentic Commerce10:45 AM · Track 3ALPHALAB: Autonomous Multi-Agent Research Across Optimization Domains with Frontier LLMsBrendan RappazzoAI in Finance10:45 AM · Track 4State of the Union: Why Local, Why NowNader Khalil, Joseph Nelson, Alex Cheema, Ahmad Osman, Matthew BermanLocal AI10:45 AM · Track 5CrabRAG: Why Automated Assistants Need Graph Memory, Not More TokensStephen ChinGraphs10:45 AM · Track 6GTM Engineering: The Technical BitsEverett BerryAI in GTM10:45 AM · Track 7From Ambient Documentation to Clinical IntelligenceChaitanya AsawaAI in Healthcare10:45 AM · Track 8DeepSWE: expert code datasetsSerena GeAgentic Engineering10:45 AM · Track 9Operating Distributed Inference Systems at ScaleNishant Gupta, Naman AhujaInference10:45 AM · Track MDiagnosing agent failures in productionPamela FoxTrack M11:00 AM · Leadership LoungeThe Agentic Product Development OrganizationMartin Harrysson, Matt Linderman, Prakhar DixitCTO Circle11:10 AM · Expo Stage 1 NEBeyond RAG: See a relational context engine reduce token burnPeter Werry11:10 AM · Expo Stage 2 NWARIA, how we built autoresearch with autoresearchZubin Aysola11:10 AM · Expo Stage 3 SWThe Lethal Trifecta Is Already on Your Developers' LaptopsMichael Patterson11:10 AM · Expo Stage 4 SEYour AI Agent Has No Nervous SystemMatt Gibiec11:10 AM · Leadership 1Tribal Dungeons of Global Shipping: AI Agents at Global ScaleDmitry BuykinAI-Native Enterprises11:10 AM · Leadership 2FinOps for AI Agents: Who Spent All the Tokens?Tisha Chawla, Susheem KoulAI Architects: AI Factories11:10 AM · Main StageMCPs, CLIs, and Skills: Choosing the Right Tooling Layer for Agentic DevelopmentNikita KothariAgentic Engineering11:10 AM · Track 1HTML Is All Agents NeedJames RussoGenerative Media11:10 AM · Track 2Why Your AI Agent Needs a Wallet: Agentic commerce on Arc with USDC and NanopaymentsHarshal BhangaleAgentic Commerce11:10 AM · Track 3Why Off-the-Shelf AI Doesn't Understand MoneyUdi MenkesAI in Finance11:10 AM · Track 4State of the Union: Why Local, Why NowNader Khalil, Joseph Nelson, Alex Cheema, Ahmad Osman, Matthew BermanLocal AI11:10 AM · Track 5Active Graph Agent Runtime (BabyAGI 4)Yohei NakajimaGraphs11:10 AM · Track 6Reverse-Engineering the AI BuyerAliisa RosenthalAI in GTM11:10 AM · Track 7Guardrails First: Engineering Member-Facing Health AIRashi AgrawalAI in Healthcare11:10 AM · Track 8Anthropic's CCA Exam as a Field-Guide for Agentic EngineeringFrank CoyleAgentic Engineering11:10 AM · Track 9Routing LLM Inference in Production: From Engine Signals to PolicyQianru Lao, Lu ZhangInference11:10 AM · Track MTracing and debugging agents across systems with OpenTelemetryChang LiuTrack M11:40 AM · Expo Stage 1 NEThe Art of Building Verifiers for Computer Use AgentsMiguel González Fernández, Corby Rosset11:40 AM · Expo Stage 2 NWSeeing the Plumbing: Profiling vLLM Speculative Decoding on NVIDIA BlackwellSheilah Kirui11:40 AM · Expo Stage 3 SWVoice is the universal interfaceKwindla Kramer, Neil Zeghidour11:40 AM · Expo Stage 4 SEThe Next Run Should Be BetterJake Broekhuizen11:40 AM · Leadership 1All the Things We Have to Do to Satisfy Your Insatiable Need for TokensDaniel Kim, Michelle NguyenInference11:40 AM · Leadership 2What If Your Chip Design Team Moved Like a Single Body?Khaled Alashmouny, Abduallah MohamedAI Architects: AI Factories11:40 AM · Main StageAuth for Agents: Unblock Autonomous AI with auth.mdMichael GrinichAgentic Engineering11:40 AM · Track 1Building an Agentic Video Editor for Mass ConsumerEkaterina DeynekaGenerative Media11:40 AM · Track 2When AI Agents Pay and Sellers Monetize: Building x402 Apps for Agentic Commerce on AWSAnil NadimintiAgentic Commerce11:40 AM · Track 3Let's integrate AI Agents in Event-Sourced SystemsDivakar KumarAI in Finance11:40 AM · Track 4Demo: GLM 5.2 on DGX Station — Frontier Intelligence Under Your DeskAhmad OsmanLocal AI11:40 AM · Track 5Your Moat Is Your Data ModelMike PhippsGraphs11:40 AM · Track 6AI in GTM at NotionFlora LiuAI in GTM11:40 AM · Track 7Shipping AI to a Million Patients Without an A/B TestJared JoselowitzAI in Healthcare11:40 AM · Track 8Guide, Verify, Solve: The Engineering Discipline Agentic Development DemandsAnirban ChatterjeeAgentic Engineering11:40 AM · Track 9Are LLM Performance Benchmarks Reliable?Ashok Chandrasekar, Jason KrambergerInference11:40 AM · Track MBenchmarking VS Code with VSC-Bench: How to measure agent performanceRoss WollmanTrack M12:05 PM · Expo Stage 1 NEThe Missing Layer in Agentic AIGiedrius Steimantas12:05 PM · Expo Stage 2 NWWhile You Were Generating: The Verification Gap Nobody Talked AboutAli Adl-Tabatabai12:05 PM · Expo Stage 3 SWMove fast and (don’t) break thingsBen Dicken12:05 PM · Expo Stage 4 SEAgents That Forge Their Own Tools: Self-Modifying AI in the WildSandhya Subramani12:05 PM · Leadership 1Stop Model Shopping: Why Ownership Beats Choice in the Agent StackPranay BhatiaInference12:05 PM · Leadership 2Preferences > Benchmarks: Model Routing for How Teams Actually BuildArchana Kamath, Tyler GillamAI Architects: AI Factories12:05 PM · Main StageHarness Engineering: Building the Production Cage for Powerful Domain AgentsMike ChambersHarness Engineering12:05 PM · Track 1The Next Game Engine Won't Have a ManualArturo NunezGenerative Media12:05 PM · Track 2x402 isn’t good (yet)Jan CurnAgentic Commerce12:05 PM · Track 3How Kepler Built Verifiable AI for Financial ServicesVinoo GaneshAI in Finance12:05 PM · Track 4Local AI DemosLocal AI12:05 PM · Track 5From Systems of Record to Systems of ContextOmri BruchimGraphs12:05 PM · Track 6The Building Blocks of GTM OrchestrationArman VaziriAI in GTM12:05 PM · Track 7200 Million Patient Interactions Later: What the Generic Voice Stack MissesVivek MuppallaAI in Healthcare12:05 PM · Track 8Benchmarking Coding Agents on New vs Legacy Code basesDenys LinkovAgentic Engineering12:05 PM · Track 9Vertical Mobility: Building an AI Inference Platform That Scales from MVP to Trillion-Parameter WorkloadsRita Zhang, Sitanshu GuptaInference12:05 PM · Track MDesign multi-agent systems that actually workTina ManghnaniTrack M12:30 PM · Expo Stage 2 NWLatent Space Live: the Inference Inflection from First Principlesswyx swyx, Rob WachenExpo Stage 21:30 PM · Expo Stage 1 NETrust, But Verify: Human-in-the-Loop for Agents That Actually MatterMichael Liendo1:30 PM · Expo Stage 2 NWYOLO Mode, Safely: microVM Sandboxes for Any AgentRowan Christmas1:30 PM · Expo Stage 3 SWYour Model is Private. Your System Isn't.Joshua Mo1:30 PM · Expo Stage 4 SEVideo Discovery for Agentic World-Model TrainingRafael Levi1:30 PM · Leadership 1From Zero to AI-Native: Scaling AI Across the OrgJosh LeavittAI-Native Enterprises1:30 PM · Leadership 2Coding Agents Don't Scale Themselves. Neither Do Your Teams.The Rise of Agent Enablement.Patrick DeboisAI Architects: AI Factories1:30 PM · Main StageLoophole - Adversarial Agents To Stress Test Your MoralityBrendan RappazzoHarness Engineering1:30 PM · Track 1While my guitar gently speaksTodd FisherGenerative Media1:30 PM · Track 2Agent Spending Without Controls: The Missing Infrastructure Layer for AI Pa…Rodrigo Coelho, Pranav MaheshwariAgentic Commerce1:30 PM · Track 3Build for the Memo, Not the Demo — Notes from 200 Investment CommitteesShawn ChanAI in Finance1:30 PM · Track 4Local Models: Trust, Control, OptimizationCarter Abdallah, Vincent Weisser, Lucas Atkins, Chris AlexiukLocal AI1:30 PM · Track 5AI : Learned Execution Graphs for Real-Time Anomaly Detection & Drift Classification in APIsRitvik PandyaGraphs1:30 PM · Track 6How Juries and Librarians Can Solve GTM's AI Trust ProblemAlex BauerAI in GTM1:30 PM · Track 7Al is becoming the World's largest Relationship Therapist. We Can't Afford to Get it Wrong.Clay Cockrell, Tony FabrikantAI in Healthcare1:30 PM · Track 8Codex, Behind the HarnessDominik KundelAgentic Engineering1:30 PM · Track 9What's New in Inference EngineeringPhilip KielyInference1:30 PM · Track MEvaluating and optimizing AI agents: from observability to continuous improvementChang LiuTrack M1:55 PM · Expo Stage 2 NWMCP doesn’t suck — your agent doesJan Curn1:55 PM · Expo Stage 4 SEEveryone talks about document search, but what about results?George He1:55 PM · Leadership 1Which AI startups actually land enterprise contracts? Lessons from evaluating 100+ AI startups at Millennium ManagementBrian LewisAI-Native Enterprises1:55 PM · Leadership 2Agent Frameworks Considered HarmfulRémi LoufHarness Engineering1:55 PM · Main Stage🎵 Every step you take, every call you make - the reliable agent stackGiselle van DongenHarness Engineering1:55 PM · Track 1Voice agents with Realtime VideoLina ColucciGenerative Media1:55 PM · Track 2Teaching agents to payAnna SpyszAgentic Commerce1:55 PM · Track 3We Vetted 2,000 AI Skills Before They Reached DevelopersLucas PalmaAI in Finance1:55 PM · Track 4Local Models: Trust, Control, OptimizationCarter Abdallah, Vincent Weisser, Lucas Atkins, Chris AlexiukLocal AI1:55 PM · Track 5Why Agentic Systems Need OntologiesFrank CoyleGraphs1:55 PM · Track 6How We Got LLMs to Recommend Our Open Source Library (Without Paying or Plug-ins)Christopher BurnsAI in GTM1:55 PM · Track 7Healthcare’s Agent Bytecode: X12 as the Harness for AI AgentsVasant KearneyAI in Healthcare1:55 PM · Track 8Multiplayer agentic engineering: enabling your whole team and your best agents to work togetherArjun SinghAgentic Engineering1:55 PM · Track 9Large clusters for small modelsDaniel SvonavaInference1:55 PM · Track MBlast Radius Zero: One‑Command OpenClaw Sandboxes in the CloudArun SekharTrack M2:00 PM · OpenAI BoothAMA with Theo Browne (@t3.gg)Theo BrowneBooth Activation2:25 PM · Expo Stage 1 NEOptimizing Open Models for Production Grade InferenceSujee Maniyam, Dylan Bristot2:25 PM · Expo Stage 2 NWImproving AIE: Q&A and Feedback Sessionswyx swyxExpo Stage 22:25 PM · Expo Stage 3 SWThe Human Is an Async APIMelanie Warrick2:25 PM · Leadership 1Your Hero Agent Needs a PartyKunal LanjewarAI-Native Enterprises2:25 PM · Main StageWe let an AI agent execute Bash and lived to talk about itSarah SandersHarness Engineering2:25 PM · Track 1Generative Video at the Speed of LightKeegan McCallumGenerative Media2:25 PM · Track 2The Agentic Commerce StackAhnaf PrioAgentic Commerce2:25 PM · Track 3Your Finance Agent's Bottleneck Is YouRamana Siddanth EmaniAI in Finance2:25 PM · Track 4Compression at the EdgeChris Alexiuk, Daniel Han, Asma Beevi, Merve Noyan, Parth SareenLocal AI2:25 PM · Track 5Video Has No Memory. Here's How We Built One.James LeGraphs2:25 PM · Track 6Knowledge Systems: The New GTM StackJeffrey WangAI in GTM2:25 PM · Track 7Trading Desks to Clinical Trials: Parallels in Applied Vertical AIAyush BhardwajAI in Healthcare2:25 PM · Track 8Always-on agents run production without the on-call taxJustin SmithAgentic Engineering2:25 PM · Track 9The Frontier AI Inference Cloud for AgentsByung-Gon (Gon) ChunInference2:25 PM · Track MOperate agents safely at scale with enterprise governanceAshu JoshiTrack M2:50 PM · Expo Stage 1 NEHarness Engineering: The New Core Skill for Agentic DevelopersDru Knox2:50 PM · Expo Stage 3 SWSmall Claws Are Beautiful: Edge Agents with NanoClaw, Raspberry Pi, and Graph MemoryJeremy AdamsExpo Stage 32:50 PM · Expo Stage 4 SEThe Software Factory2:50 PM · Leadership 1AI Agents Are Just Distributed Systems NowSalman MunafAI-Native Enterprises2:50 PM · Leadership 2Inside 847 Production Clinical AI NotesSebastian FoxAI Architects: AI Factories2:50 PM · Main StageNo Memory, No Harness: Why the Database Is the Last Line of DefenseKay MalcolmHarness Engineering2:50 PM · Track 1Infra behind Krea 2 - How to train and serve at scaleGabriel Jorge MenezesGenerative Media2:50 PM · Track 2Your Agent Just Authorized What?!Jay MokAgentic Commerce2:50 PM · Track 3Simulation-Maxxing: How Nubank ships agents 20× faster with simulationsShreya Rajpal, Aman GuptaAI in Finance2:50 PM · Track 4Compression at the EdgeChris Alexiuk, Daniel Han, Asma Beevi, Merve Noyan, Parth SareenLocal AI2:50 PM · Track 5On-Device Agentic AI for the New York Times GamesShafik Quoraishee, Joanne SongGraphs2:50 PM · Track 6How AI Agents Let GTM Teams ScaleJustin JoyceAI in GTM2:50 PM · Track 7How to build an AI-Native Health CompanyDan FengAI in Healthcare2:50 PM · Track 8Realtime multiplayer, automation, and you!Idan GazitAgentic Engineering2:50 PM · Track 9KV Cache-Aware Routing and P/D Disaggregation on Kubernetes: The Parts Public Benchmarks Don't ShowYuchen Fama, Ashish KamraInference3:20 PM · Expo Stage 1 NEAgent Memory Is a Solved Problem. Agent Learning Is Not.Karthik Ranganathan, Heather Downing3:20 PM · Expo Stage 2 NWImproving AIE: Q&A and Feedback Session (repeat)swyx swyxExpo Stage 23:20 PM · Expo Stage 3 SWAn Interaction Is All You NeedIvan LeoExpo Stage 33:20 PM · Expo Stage 4 SEAn AI Future Without the Lock-InRemy Guercio3:20 PM · Leadership 1The Signal Layer: What to Build When Anything Can Be BuiltLena HallAI-Native Enterprises3:20 PM · Leadership 2Give the Agent a Budget, Not a TokenSachin MalhotraAI Architects: AI Factories3:20 PM · Main StageHow we Solved Agent BuildingAndrew QuHarness Engineering3:20 PM · Track 1The Next Medium: Why Real-Time Interactive Video Changes Everything for DevelopersAhmed AhresGenerative Media3:20 PM · Track 2The End of the Static Screen: Architecting Intent-Driven UX with Agentic OrchestrationGus IwanagaAgentic Commerce3:20 PM · Track 3Skills are new features: Building Skill-Centric Harness for Agentic ProductsYogendra MirajeAI in Finance3:20 PM · Track 4Model RoutingNader Khalil, Walden Yan, Tanay Varshney, Alex AtallahLocal AI3:20 PM · Track 5Citation Needed: Provenance for LLM-Built Knowledge GraphsDaniel ChalefGraphs3:20 PM · Track 6Building GTM AI Agents: Lessons from Deploying to 6,000 UsersSait IzmitAI in GTM3:20 PM · Track 7Don't be data poorAnuj IravaneAI in Healthcare3:20 PM · Track 8Velocity Sickness: What Happens When Your Whole Team Gets 10x FasterMatt DaileyAgentic Engineering3:20 PM · Track 9Two Bugs That Hid in Plain Sight: A vLLM Debugging Detective StoryAsaf Gardin, Yuval BelferInference3:45 PM · Expo Stage 1 NETaking Reinforcement Learning Cross DatacenterNan Jiang3:45 PM · Expo Stage 2 NWDashboards are DeadSarah Simionescu3:45 PM · Leadership 1Tell the Robot What You WantSandhya SubramaniAI-Native Enterprises3:45 PM · Main StageAgents Without Code: How Skills, YAML, and Filesystems Replaced PythonPhilipp SchmidHarness Engineering3:45 PM · Track 2Beyond the Lethal Trifecta: Agentic Commerce on the Open Internet at Machine SpeedDavid LevineAgentic Commerce3:45 PM · Track 3Wearing the Agent: Engineering a Family-and-Friends Personal Agent, from Group Chats to GlassesSai Krishna RallabandiAI in Finance3:45 PM · Track 4Model RoutingNader Khalil, Walden Yan, Tanay Varshney, Alex AtallahLocal AI3:45 PM · Track 5Why We Killed Our Multi-Agent Pipeline: Lessons From Pharma Commercial IntelligenceSubbiah Sethuraman, Abhilash AsokanGraphs3:45 PM · Track 6The Death of Developer AdvocatesStephanie JarmakAI in GTM3:45 PM · Track 7Why Your Enterprise Tech Stack Isn't Ready for AI Agents - And What to Build InsteadChristopher Lovejoy, Saul HowardAI in Healthcare3:45 PM · Track 8Open Source Is Dead. Long Live Open Source.Saoud RizwanAgentic Engineering3:45 PM · Track 9Weight Folding, CUDA Streams, and the Bug That Made My Model Speak BackwardsFilip MakraduliInference4:30 PM · Main StageClosing Keynote — Theo BrowneTheo BrowneMain Stage4:50 PM · Main StageClosing Keynote: Garry TanGarry TanMain Stage5:10 PM · Main StageStartup BattlefieldHowie Liu, Joshua Xu, swyx swyxMain Stage