两天议程，
一目了然。

Opening with swyx · Video

From Models to Working Enterprise Agents

Building Agents: Yesterday, Today, Tomorrow

When we can do anything, do we still know what we want?

Post-Training at the Frontier: Teaching Open Models to Search and Browse the Web

Full-Stack Inference Engineering: Latency, Throughput and Cost

Measure the frontier of AI for real world use

Community in the AI era

From Model Intelligence to Agent Harness: Enabling Long-Horizon Agents

Dark Factory to Lobsterville: Intent Engineering for Persistent Multiplayer Agents

On-Device AI: Bringing Intelligence Closer to Everyone

Kubernetes Is Not Your Sandbox

AI-Native Work at Manus: One Monorepo, Open Tasks

Powering open source personal agents with Exa search

Your Agent's Memory Is Next Quarter's Training Data

Building a reliable, multi-region LLM gateway

Nobody Planned AI Infra: How AI Builders Pushed a Database Beyond the Database

Optimizing Organization-Wide LLM Inference Cost: Seat License, Pay-per-Token, or Provisioned, and Beyond

Reinforcement Learning at Scale on Agent Substrate: Orchestrating Millions of Sandboxes

Life of a Game Developer in AI era

Building a Personal Agent for Growth with YouMind

Trust Your Own Evals, Not the Leaderboard: What 10k+ Internal Agent PRs Taught Us About Adopting Open Models

Beyond the Pilot: Managing Your 500th AI Agent

Beyond Reasoning: Building Infrastructure for AI Agents That Act

Beyond Screenshots: Building Agent-Native Verification for Mobile Apps

Agentic Harness: Real-Time Audio and Video AI Interviews

Your Repo. Your Benchmark.

Trusted Context: Building the Layer Between Enterprise Data and AI Agents

The Harness Is the First Reviewer

Your Data Agent Needs More Than a Knowledge Graph -- From Semantic Layers to Executable Ontologies with Apache Ossie

From Idea to Viral Launch: How I Built a 3D Little Prince Game with AI

From Matching to Delivery: Scaling Creator Partnerships with AI

Taxonomy of World Models

HTML is all your agents need: Redefining Context Engineering

AGI is decoding. Encoding is the moat.

AI in Manufacturing: The Last Mile Is Half the Journey

Trusting Evals I Can’t Read

The Good, the Bad and the Ugly: What 3,000 live sessions taught us about streaming speech-to-text

AI startups: products, markets & what comes next

Stronger Models, Leaner Harnesses: What to Control and What to Let Go

Beyond Vision-Language-Action: First Principles for Building Physical AI

From 3D Generation to World Models: Building Persistent Interactive Worlds for Physical AI

How We're Building Open Decision Models for On-Device Agents

Communication Among Sovereign Agents: The Foundation of Agent Economy

Real2Real: A Quantum Embodied AI Paradigm for Optimal Behaviors Generation

Noe-0: A World Action Model Built on Real-World Human Experience

Building Robot Hardware for Rapidly Evolving Models

Talk to be announced

From Hardware to Intelligence: Full-Stack Evolution and Real-World Deployment of Home Service Robots

Robot Data Needs a Build System

Memory Engineering, RSI and Team Knowledge Layer

Reading the Web at Trillion-Token Scale