> Markdown version of [/videos/world-congress-2026-north-america?page=12](https://www.wearedevelopers.com/videos/world-congress-2026-north-america?page=12). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Videos 2651 videos · 28490 moments · 1454.1 hours of content ## World Congress North America 2026 videos ### From Software Agents to Physical Devices: Inside the Agentic Hardware Stack AI coding agents are maxing out your attention budget. Discover how purpose-built agentic hardware uses Rust and MQTT to securely orchestrate parallel AI workflows away from distracting screens. - **Speakers:** Michael Yuan, Vivian Hu - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 29:30 - [Watch](https://www.wearedevelopers.com/videos/100454-from-software-agents-to-physical-devices-inside-the-agentic-hardware-stack) ### Context Engineering Kung Fu Large language models suffer from anterograde amnesia, making context engineering the true battleground for agent reliability. Master advanced state management and deferred tool loading to eliminate costly context rot. - **Speakers:** Carl Lapierre - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 30:51 - [Watch](https://www.wearedevelopers.com/videos/100439-context-engineering-kung-fu) ### Who Tests the AI? Building Trustworthy AI Systems at Enterprise Scale AI doesn't need to crash to fail—a convincing hallucination can instantly destroy customer trust. Master continuous evaluation frameworks to secure probabilistic systems at enterprise scale. - **Speakers:** Him Raj Singh - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 27:57 - [Watch](https://www.wearedevelopers.com/videos/100413-who-tests-the-ai-building-trustworthy-ai-systems-at-enterprise-scale) ### File Systems Are the New Primitive for AI Agents File systems are the missing foundational primitive for AI agents. Ditch ephemeral chat windows for directory structures to build durable, inspectable, and highly collaborative workflows in production. - **Speakers:** Andrew Wong - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 22:37 - [Watch](https://www.wearedevelopers.com/videos/100392-file-systems-are-the-new-primitive-for-ai-agents) ### Shipping AI Features People Actually Use Generating 99.5% accurate AI results is easy, but bridging that final 0.5% separates gimmicks from winners. Learn to move past demo hype and build features users actually adopt. - **Speakers:** Ben Kus, Stefan Lederer, Wayne Liu, Floriane Gramlich - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 29:36 - [Watch](https://www.wearedevelopers.com/videos/100611-shipping-ai-features-people-actually-use) ### Run your agents in Kubernetes: Build once, deploy anywhere. But really? An agentic LLM app ran flawlessly on a local Mac but failed instantly on AWS EKS. Learn how to overcome architecture mismatches, storage errors, and resource starvation in Kubernetes. - **Speakers:** Michal Salanci - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 21:26 - [Watch](https://www.wearedevelopers.com/videos/100584-run-your-agents-in-kubernetes-build-once-deploy-anywhere-but-really) ### Closing the autonomous software development loop Autonomous coding agents fly blind in production without Agentic Session Review. Learn how empowering AI to watch real user journeys transforms reactive bug fixing into proactive application optimization. - **Speakers:** Justin Halsall, Scott Voigt - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 26:49 - [Watch](https://www.wearedevelopers.com/videos/100522-closing-the-autonomous-software-development-loop) ### Sandboxing the Swarm: Building Secure, Serverless AI Agents with Wasm Stop relying on complex container hardening for autonomous AI. WebAssembly introduces a zero-trust, default-deny sandbox to securely execute dynamically generated agent code with sub-millisecond cold starts. - **Speakers:** Thorsten Hans - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 29:16 - [Watch](https://www.wearedevelopers.com/videos/100482-sandboxing-the-swarm-building-secure-serverless-ai-agents-with-wasm) ### From Cloud Native to Multi-Cloud Native: Write Once, Deploy Anywhere Are you truly multi-cloud, or just vendor-locked with containers? Discover how driver-based abstractions decouple business logic from proprietary SDKs. Write code once, and deploy anywhere. - **Speakers:** Sandeep Pal - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 26:32 - [Watch](https://www.wearedevelopers.com/videos/100458-from-cloud-native-to-multi-cloud-native-write-once-deploy-anywhere) ### From Stateless to Self-Improving: Building Agent Workflows That Get Better Every Session Tired of re-teaching your AI assistant the same codebase rules? Learn to build persistent, self-improving agent workflows that retain memory, master new skills, and evolve across sessions. - **Speakers:** Laura Salinas - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 28:44 - [Watch](https://www.wearedevelopers.com/videos/100429-from-stateless-to-self-improving-building-agent-workflows-that-get-better-every-session) ### Fast, Cheap, and Accurate: Optimizing LLM Inference with vLLM and Quantization Struggling to balance cost, accuracy, and latency in your LLM deployments? Discover how vLLM and quantization can slash your VRAM needs by 50% without sacrificing performance. - **Speakers:** Legare Kerrison, Cedric Clyburn - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 26:27 - [Watch](https://www.wearedevelopers.com/videos/100416-fast-cheap-and-accurate-optimizing-llm-inference-with-vllm-and-quantization) ### Your Threat Model Is Lying to You: Why Modeling the Design Isn’t Enough in 2026 Your fifteen declared dependencies just ballooned into hundreds of hidden vulnerabilities. Discover why design-phase threat modeling fails today and how continuous feedback loops expose your true production risks. - **Speakers:** Farshad Abasi - **Event:** World Congress 2026 North America - **Published:** September 25, 2026 - **Duration:** 30:06 - [Watch](https://www.wearedevelopers.com/videos/100396-your-threat-model-is-lying-to-you-why-modeling-the-design-isn-t-enough-in-2026) [Next page](https://www.wearedevelopers.com/videos/world-congress-2026-north-america?page=13)