[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6},"actionData","errors","routes/public/w-agenda",{"_7":8,"_9":10,"_11":-5,"_12":13,"_14":15,"_16":17,"_18":19,"_20":21,"_22":23,"_24":25,"_26":27,"_28":8,"_29":8},"published",true,"event",{"_253":330,"_331":332,"_333":334,"_60":335,"_62":336,"_337":338,"_339":340},"accent","days",[317,318,319],"day","2026-09-16","dayLongLabel","Wednesday, September 16","window",{"_311":312,"_313":314,"_315":316},"rooms",[59,89,122,187],"blocks",[258,259,260,261,262,263,264,265,266,267,268,269],"legend",[250,251,252],"daySessions",[30,31,32,33,34,35,36,37,38,39,40,41],"scheduledAny","anyPublicSessions",{"_42":235,"_44":236,"_46":237,"_48":238,"_50":51,"_52":53,"_54":55,"_56":57,"_58":59,"_60":239,"_62":240,"_64":241},{"_42":222,"_44":223,"_46":224,"_48":225,"_50":105,"_52":106,"_54":87,"_56":88,"_58":59,"_60":188,"_62":212,"_64":226},{"_42":208,"_44":209,"_46":210,"_48":211,"_50":105,"_52":106,"_54":87,"_56":88,"_58":89,"_60":188,"_62":212,"_64":213},{"_42":181,"_44":182,"_46":183,"_48":184,"_50":51,"_52":53,"_54":185,"_56":186,"_58":187,"_60":188,"_62":189,"_64":190},{"_42":175,"_44":176,"_46":177,"_48":178,"_50":105,"_52":106,"_54":87,"_56":88,"_58":59,"_60":151,"_62":152,"_64":179},{"_42":162,"_44":163,"_46":164,"_48":165,"_50":105,"_52":106,"_54":87,"_56":88,"_58":89,"_60":151,"_62":152,"_64":166},{"_42":147,"_44":148,"_46":149,"_48":150,"_50":120,"_52":121,"_54":87,"_56":88,"_58":122,"_60":151,"_62":152,"_64":153},{"_42":134,"_44":135,"_46":136,"_48":137,"_50":120,"_52":121,"_54":87,"_56":88,"_58":59,"_60":123,"_62":124,"_64":138},{"_42":116,"_44":117,"_46":118,"_48":119,"_50":120,"_52":121,"_54":87,"_56":88,"_58":122,"_60":123,"_62":124,"_64":125},{"_42":101,"_44":102,"_46":103,"_48":104,"_50":105,"_52":106,"_54":87,"_56":88,"_58":59,"_60":90,"_62":91,"_64":107},{"_42":83,"_44":84,"_46":85,"_48":86,"_50":51,"_52":53,"_54":87,"_56":88,"_58":89,"_60":90,"_62":91,"_64":92},{"_42":43,"_44":45,"_46":47,"_48":49,"_50":51,"_52":53,"_54":55,"_56":57,"_58":59,"_60":61,"_62":63,"_64":65},"id","demo-s-48","friendlyNum",49,"title","Closing Keynote: The Boring Future of AI Engineering","descriptionHtml","

A bet on where this goes: fewer demos, more contracts; fewer frameworks, more schemas; and the unglamorous integration work that will decide which products are still running in five years.

","trackId","demo-track-0","trackName","AI Engineering","formatId","demo-format-0","formatName","Keynote (45 min)","roomName","Main Stage","startsAt",1789596900000,"endsAt",1789599600000,"speakers",[66],{"_67":68,"_69":70,"_71":72,"_46":73,"_74":75,"_76":77,"_78":79,"_80":81,"_82":8},"personId","demo-p-17","firstName","Ben","lastName","Oyelaran","Principal Architect","company","Steelhead","bio","Ben designs integration platforms for companies with more legacy systems than engineers. He believes most AI adoption problems are really data-contract problems wearing a new hat.","photoFileId","demo-photo-17","role","speaker","isPrimary","demo-s-47",48,"The Support Queue Is Your Best Eval Set","

Every complaint is a labelled failure somebody already wrote down for you. How we turn support tickets into evaluation cases weekly, and what that changed about which bugs we fix first.

","demo-format-1","Talk (30 min)","Room 2A",1789593300000,1789595100000,[93],{"_67":94,"_69":95,"_71":96,"_46":97,"_74":98,"_76":99,"_78":100,"_80":81,"_82":8},"demo-p-22","Leila","Farsi","Staff Product Manager","Coastline","Leila owns the trust and safety surface for Coastline's assistant. She used to be a reporter, which shows in the way she writes specifications.","demo-photo-22","demo-s-46",47,"Cutting Our Inference Bill in Half Without Touching Quality","

Batching, caching and routing decisions that halved our spend, plus the two optimisations that looked great on a dashboard and cost us accuracy. Numbers from a fleet running around the clock.

","demo-track-1","Platform & Infra",[108],{"_67":109,"_69":110,"_71":111,"_46":112,"_74":113,"_76":114,"_78":115,"_80":81,"_82":8},"demo-p-9","Felix","Braun","Infrastructure Lead","Morgenrot","Felix keeps Morgenrot's GPU fleet busy and its finance team calm. He wrote two internal post-mortems that are now required onboarding reading.","demo-photo-9","demo-s-32",33,"Deleting Internal Tools: A Year of Saying No","

We removed nineteen internal tools in twelve months and replaced them with four. How we chose, how we migrated people who did not want to move, and the two deletions we regret.

","demo-track-2","Developer Experience","Room 2B",1789589700000,1789591500000,[126],{"_67":127,"_69":128,"_71":129,"_46":130,"_74":131,"_76":132,"_78":133,"_80":81,"_82":8},"demo-p-24","Zoe","Kimura","Engineering Manager","Fernbrook","Zoe leads the platform team at Fernbrook and spent the past year replacing internal tools with fewer, better ones. She keeps a public list of everything she deleted.","demo-photo-24","demo-s-28",29,"Growing an Engineering Org Around an AI Product","

We went from twelve engineers to ninety while shipping our first AI product, and got a lot of it wrong. Hiring, team topology, and the two reorganisations we could have skipped.

",[139],{"_67":140,"_69":141,"_71":142,"_46":143,"_74":144,"_76":145,"_78":146,"_80":81,"_82":8},"demo-p-20","Aisha","Rahman","VP Engineering","Loomis","Aisha grew Loomis engineering from twelve people to ninety while shipping the company's first AI product. She writes mostly about the parts of that which went badly.","demo-photo-20","demo-s-22",23,"Undo Is a Feature: Designing Reversibility Into AI Products","

Confidence in an AI product comes from being able to take it back. A design tour of undo, diffing and preview patterns, and the data model changes each one demands.

",1789582500000,1789584300000,[154],{"_67":155,"_69":156,"_71":157,"_46":158,"_74":159,"_76":160,"_78":161,"_80":81,"_82":8},"demo-p-14","Ines","Moreau","Design Lead","Calyx","Ines designed Calyx's assistant interface and then its undo system, in that order of difficulty. She is a recovering print designer who still sets type by hand for fun.","demo-photo-14","demo-s-27",28,"Scheduling GPU Work Without a Scheduler Team","

Fair queuing, preemption and priority for a fleet you cannot afford to leave idle — implemented by two engineers with a job queue and a lot of restraint about what to build.

",[167],{"_67":168,"_69":169,"_71":170,"_46":171,"_74":172,"_76":173,"_78":174,"_80":81,"_82":8},"demo-p-19","Theo","Whitfield","Staff Engineer","Junction","Theo works on Junction's orchestration engine and the scheduler underneath it. He came to distributed systems from computational biology and has never entirely left.","demo-photo-19","demo-s-25",26,"Your Legacy Systems Are the Hard Part","

Most AI integration projects stall on a mainframe export, not a model. Patterns for wrapping systems you cannot change, and how to negotiate data contracts with teams that have no incentive to help you.

",[180],{"_67":68,"_69":70,"_71":72,"_46":73,"_74":75,"_76":77,"_78":79,"_80":81,"_82":8},"demo-s-5",6,"Evals Are Your Only Moat: Building Test Suites for LLM Products","

Model upgrades break prompts silently. In this hands-on workshop you will build an eval pipeline from scratch: rubric design, failure taxonomies, CI wiring, and the regression report that doubles as your product spec. Bring a laptop.

","demo-format-3","Workshop (120 min)","Workshop Lab",1789579800000,1789587000000,[191,192],{"_67":201,"_69":202,"_71":203,"_46":204,"_74":205,"_76":206,"_78":207,"_80":81,"_82":8},{"_67":193,"_69":194,"_71":195,"_46":196,"_74":197,"_76":198,"_78":199,"_80":81,"_82":200},"demo-p-16","Hana","Sato","Research Engineer","Aperture AI","Hana works on small-model distillation and spends most of her week reading evaluation traces. She was a computational linguist before that and it shows in her error taxonomies.","demo-photo-16",false,"demo-p-5","Tom","Achebe","CTO","Forgeline","Tom co-founded Forgeline after leading platform engineering at two payments companies. He is the reason Forgeline had an evaluation suite before it had a marketing site.","demo-photo-5","demo-s-11",12,"Postgres to the Edge: A Migration Post-Mortem","

What broke, what got faster, and the three queries that nearly made us roll the whole thing back. An honest accounting with p99 charts and the cost line nobody put in the original proposal.

",1789581600000,[214],{"_67":215,"_69":216,"_71":217,"_46":218,"_74":219,"_76":220,"_78":221,"_80":81,"_82":8},"demo-p-11","Omar","Haddad","Distinguished Engineer","Qanat Systems","Omar has spent twenty years on data infrastructure, most recently the query layer behind Qanat's edge platform. He is unreasonably interested in what happens at p99.9.","demo-photo-11","demo-s-20",21,"On Call for a Model: Reliability Practices for Inference Platforms","

Inference platforms fail in ways your existing runbooks do not describe. We rebuilt our alerting around output quality rather than uptime, and the pager got quieter and more useful at the same time.

",[227],{"_67":228,"_69":229,"_71":230,"_46":231,"_74":232,"_76":233,"_78":234,"_80":81,"_82":8},"demo-p-12","Nina","Berg","Staff SRE","Halberd","Nina is on call for a platform that serves inference in forty countries. She thinks most reliability work is really vocabulary work: naming failure modes precisely enough to fix them.","demo-photo-12","demo-s-21",22,"Human in the Loop, For Real: Clinical AI Under Review","

Every model output in our product is read by a clinician before it reaches a patient chart. This keynote covers what that constraint taught us about interface design, latency budgets, and the difference between assistance and automation.

",1789576200000,1789578900000,[242],{"_67":243,"_69":244,"_71":245,"_46":246,"_74":247,"_76":248,"_78":249,"_80":81,"_82":8},"demo-p-13","Kwame","Mensah","Director of Engineering","Northwind Health","Kwame leads the clinical AI group at Northwind, where every model output is reviewed by a clinician before it reaches a chart. He is candid about what regulation does and does not actually slow down.","demo-photo-13",{"_42":51,"_253":53,"_254":257},{"_42":105,"_253":106,"_254":256},{"_42":120,"_253":121,"_254":255},"name","tone",3,2,1,{"_42":235,"_270":59,"_46":237,"_64":308,"_272":309,"_274":310,"_254":257,"_52":53,"_276":200,"_277":288,"_279":304,"_281":282,"_283":257},{"_42":222,"_270":59,"_46":224,"_64":307,"_272":299,"_274":275,"_254":256,"_52":106,"_276":200,"_277":278,"_279":288,"_281":282,"_283":257},{"_42":175,"_270":59,"_46":177,"_64":300,"_272":290,"_274":291,"_254":256,"_52":106,"_276":200,"_277":292,"_279":288,"_281":282,"_283":257},{"_42":134,"_270":59,"_46":136,"_64":306,"_272":285,"_274":286,"_254":255,"_52":121,"_276":200,"_277":287,"_279":288,"_281":282,"_283":257},{"_42":101,"_270":59,"_46":103,"_64":305,"_272":294,"_274":295,"_254":256,"_52":106,"_276":200,"_277":296,"_279":288,"_281":282,"_283":257},{"_42":43,"_270":59,"_46":47,"_64":300,"_272":301,"_274":302,"_254":257,"_52":53,"_276":200,"_277":303,"_279":304,"_281":282,"_283":257},{"_42":208,"_270":89,"_46":210,"_64":298,"_272":299,"_274":275,"_254":256,"_52":106,"_276":200,"_277":278,"_279":288,"_281":282,"_283":257},{"_42":162,"_270":89,"_46":164,"_64":297,"_272":290,"_274":291,"_254":256,"_52":106,"_276":200,"_277":292,"_279":288,"_281":282,"_283":257},{"_42":83,"_270":89,"_46":85,"_64":293,"_272":294,"_274":295,"_254":257,"_52":53,"_276":200,"_277":296,"_279":288,"_281":282,"_283":257},{"_42":147,"_270":122,"_46":149,"_64":289,"_272":290,"_274":291,"_254":255,"_52":121,"_276":200,"_277":292,"_279":288,"_281":282,"_283":257},{"_42":116,"_270":122,"_46":118,"_64":284,"_272":285,"_274":286,"_254":255,"_52":121,"_276":200,"_277":287,"_279":288,"_281":282,"_283":257},{"_42":181,"_270":187,"_46":183,"_64":271,"_272":273,"_274":275,"_254":257,"_52":53,"_276":200,"_277":278,"_279":280,"_281":282,"_283":257},"room","Tom Achebe, Hana Sato","timeLabel","10:30 AM – 12:30 PM","startLabel","10:30 AM","conflicted","startMin",90,"minutes",120,"lane",0,"lanes","Zoe Kimura","1:15 PM – 1:45 PM","1:15 PM",255,30,"Ines Moreau","11:15 AM – 11:45 AM","11:15 AM",135,"Leila Farsi","2:15 PM – 2:45 PM","2:15 PM",315,"Theo Whitfield","Omar Haddad","10:30 AM – 11:00 AM","Ben Oyelaran","3:15 PM – 4:00 PM","3:15 PM",375,45,"Felix Braun","Aisha Rahman","Nina Berg","Kwame Mensah","9:30 AM – 10:15 AM","9:30 AM","startMs",1789574400000,"startHour",9,"endHour",16,{"_320":327,"_322":328,"_324":329},{"_320":15,"_322":326,"_324":17},{"_320":321,"_322":323,"_324":325},"key","2026-09-17","label","Thu, Sep 17","longLabel","Thursday, September 17","Wed, Sep 16","2026-09-15","Tue, Sep 15","Tuesday, September 15","AI Builders Summit 2026","slug","ai-builders-summit-2026","location","Fort Mason Center, San Francisco",1789473600000,1789646400000,"timezone","America/Los_Angeles","notice","Demo conference. The talks, speakers and reviews below are invented for demonstration — nobody named here submitted anything."]