У нас вы можете посмотреть бесплатно AI Agents Need You. Are You Ready? 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
This stream is sponsored by Factory AI. All opinions are my own. Stripe ships 1,000 PRs a week from AI agents. Cursor's team has 35% of their code written by autonomous agents. Factory AI's Droid just ran a coding mission for 16 days straight. The shift from prompting to autonomous agent systems already happened. Most developers haven't caught up yet. Today I'm sitting down with the Factory AI founding team for a deep technical conversation about Missions, their new system where you give an AI agent a goal and it figures out the rest. No prompting. No hand-holding. You send it on a mission and come back to working code. We go deep on: How Missions actually works under the hood Why they went multi-model (different models for different jobs) The self-improving Signals system (the agent detects its own mistakes and fixes them) How this compares to what Stripe and Cursor are doing What this means for engineers and the future of software development ANNOUNCEMENT: We're hosting a live build event at Factory AI HQ next week before GTC. 200 million free tokens for attendees to try Missions themselves. Sign up here: [LUMA LINK] Factory AI Missions Event 3/12: https://rfer.me/sfmissions Featuring: Eno Reyes, Co-founder, Factory AI Theo Luan, Missions Lead, Factory AI Luke Alvoeiro, Engineer, Factory AI Timestamps 00:00:00 The Coming Shift in AI Agents 00:02:33 Factory AI and Droid Platform Overview 00:05:07 Missions Inspiration: Anthropic Blog Post 00:07:38 Validation Contracts for Accurate Software 00:10:03 Live Demo: Canoe Club App Mission 00:12:17 Agent Best Practices and TDD 00:15:07 Specs vs. Planning in Agentic Engineering 00:19:25 Enterprise Needs and Building Trust in AI 00:24:30 Solving the 80/20 Coding Agent Problem 00:28:38 The Agent Readiness Framework 00:31:22 Missions High-Level Architecture 00:34:30 Multi-Model Strategy for Different Tasks 00:36:05 Continuous Learning and Agent Skills 00:40:07 Agent Architectures and Parallelization 00:46:01 Token Cost Savings with Model Agnosticism 00:48:57 Meet the Factory AI Founding Team 00:52:53 Initial Challenges and UX Optimization 00:54:45 The Compressing Context Technique 00:57:15 The Dark Factory and Software's Future 00:59:34 Luma Event & Token Giveaway Announcement 01:02:07 Internal Workflow and Dogfooding 01:07:20 Future Feature: Persistent Droid Computers 01:10:37 Good vs. Bad Use Cases for Missions 01:15:45 Inside a Worker Session: Handoffs 01:19:29 Building the Missions TUI with a Mission 01:21:42 SAS Bench: Replicating SaaS Apps with AI 01:26:52 The Evolving Role of Software Engineers 01:31:59 AI Model Personalities as 'Creatures' 01:35:09 Comparing Model Personalities: Codex vs. Opus 01:40:00 Cybernetics and Stable Feedback Loops 01:44:30 The Nuances of Human Prompt Engineering 01:48:30 Final Advice and Call for Feedback 01:49:01 Event Details and Final Wrap-Up Factory AI: https://factory.ai Missions: https://factory.ai/news/missions Follow me: X: https://x.com/RayFernando1337 Discord: https://rfer.me/discord #AIAgents #SoftwareEngineering #FactoryAI #Missions #AgenticEngineering