У нас вы можете посмотреть бесплатно Claude AI is not intelligent | Stochastic Architecture | Vaughn Vernon | SAG 2025 или скачать в максимальном доступном качестве, видео которое было загружено на ютуб. Для загрузки выберите вариант из формы ниже:
Если кнопки скачивания не
загрузились
НАЖМИТЕ ЗДЕСЬ или обновите страницу
Если возникают проблемы со скачиванием видео, пожалуйста напишите в поддержку по адресу внизу
страницы.
Спасибо за использование сервиса ClipSaver.ru
LLMs don’t reason. They don’t understand. They predict. If you’re using #chatgpt, #claude, #openclaw or any AI tool to write production code — this talk may completely change how you think about using AI responsibly in production-grade systems. In this keynote from the #iSAQB "Software Architecture Gathering 2025", renowned software architect and author Vaughn Vernon exposes the uncomfortable truth behind AI coding assistants and Large Language Models in modern software development. He cuts through the AI hype and explains what LLMs truly mean for modern software architecture. Introducing “Stochastic Architecture", Vaughn explains why LLMs are not reasoning systems but probabilistic prediction machines — optimised to sound convincing, not to be correct. They generate statistically plausible answers based on patterns in data, not verified truth or genuine understanding. Through hands-on experience building open-source tools with modern AI coding assistants, he reveals both the mind-blowing productivity gains and the subtle architectural traps that can undermine quality, correctness, and long-term maintainability. This keynote is a must-see for software architects, developers, and tech leads who want to use AI responsibly — without sacrificing engineering craft. In this keynote, you’ll discover: • Why LLMs are non-deterministic and probabilistic by design • How “hallucinations” are not bugs — but expected behavior • Why AI is optimized for pleasing you, not being correct • The architectural risks of over-trusting generated code • How context window limits impact large systems • Why AI-generated tests can silently undermine correctness • Real-world stories of AI accelerating development — and wiping out entire codebases • Why “Speed is abundant. Craft is scarce.” • How Domain-Driven Design and bounded contexts help you work with AI, not against it • Practical guardrails for safely integrating AI into professional software engineering _______ Resources mentioned in the Keynote: https://Github.com/VaughnVernon https://vaughnvernon.com _______ Save the date for iSAQB Software Architecture Gathering 2026, from November 16 to 19, 2026 in Berlin: https://sag.isaqb.org Follow us on: Instagram: / isaqb.official LinkedIn: / isaqb.org Mastodon: https://mastodon.social/@isaqb Bluesky: https://isaqb.bsky.social Facebook: https://facebook.isaqb.org Mastodon: https://mastodon.social/@isaqb _______ Video Production @Kameramensch-Filmproduktion for iSAQB GmbH Executive Producer: Johannes Klemt 1st Camera OP: Nico Schallat 2nd Camera OP: Vladimir Traut Stage, Light & Sound: Kingsize.Events