AI-generated English translations of new videos. A no-sign-up video summary & translation page
📺 Data Agents over S3: Build the Missing Semantic Layer for Files - Dmitry Petrov, DataChain
📺 Multi-Agent SRE: What Happens When Your Agents Want Opposite Things - Aditi Mewada, Apple
📺 Building an Agentic Eval Pipeline: Battle-Tested Lessons... - Kurtis Van Gent & Haoyu Wang, Google
📺 Agentic AI from Cloud to Edge: Building Intelligent Systems on Arm - Sudheera Vanguri, Arm
📺 Why Domain-specific Agents Are the Future - Justin Schroeder, Standard Agents
📺 Cut The Noise: Building a Code Reviewer You Can Trust - Joah Gerstenberg, Block, Inc
📺 The Key to Creating GREAT OpenAI Plugins- Thomas Sarlandie, OpenAI
Thomas Alandi from OpenAI's Plugins team presents practical guidance for developers building MCP servers for ChatGPT, drawing on questions from the developer community and usage data from OpenAI's public directory. The talk addresses how to structure tools, schemas, annotations, and server instructions to improve discoverability and conversation quality.
■ Tools
- How many tools to expose to the model and why there is no hard limit
- Common anti-patterns such as hiding real tools behind a generic command list
■ Schemas and Annotations
- Keeping schema sizes reasonable and reviewing them for errors
- Why annotations are required for submission and how they affect auto-approval vs. destructive-tool confirmation
■ Server Instructions and OpenAI's MCP Contributions
- Recommended length and placement of server instructions
- OpenAI's MCP apps, ChatGPT Add-Ons, event support, and file upload unification
Geared toward developers publishing MCP servers or ChatGPT plugins, the talk offers actionable best practices and insight into how OpenAI evaluates submissions. Viewers can apply these guidelines to improve their tool design and increase the chances of successful directory integration.
📺 The Context ft. Obot AI
Craig Galik and Donnie Adams from Obot demonstrate how to run self-hosted Claude environments on your own Kubernetes infrastructure, giving organizations full control over sessions, networking, secrets, and audit logging while preserving the native Claude experience. The session covers architecture, configuration, and real-world use cases for automated software workflows.
■ Self-Hosted Claude Environments
- Overview of self-hosted Claude environments built on Anthropic's hosted environment feature
- Running sessions behind your own firewall with full OS-level and network control
- Obot acting as gateway middleware for session authorization, profiles, and policies
■ Live Demo and Configuration
- Agent Substrate (Google) as the base layer for launching isolated sandboxed sessions
- LLM gateway and MCP gateway integration for audit logging and model switching
- Secrets management, managed settings, and web access controls per environment
- Per-user access profiles, cloud MCP permissions, and restricted environment comparisons
- Audit logs tracking outbound requests, blocked calls (403 responses), and session activity
■ Automated Software Factory Use Cases
- Scheduled tasks triggered by GitHub issue labels to auto-fix bugs and open pull requests
- Daily tech update tasks pulling changes from MCP-based projects and posting to Slack
- Layered workflow stages (review, QA) using GitHub project boards as a mini software factory
■ Governance and Future Direction
- Balancing security policies across sandbox isolation, network controls, and data classification
- Integrating AI into the management layer for insights across sessions, interactions, and vulnerabilities
- Access path: Obot 0.27 release notes, blog post, and documentation for Kubernetes setup
プラットフォームやゲートウェイの仕組み、セキュリティポリシーの設計、自動化ワークフローの構築に関心のあるエンジニアやセキュリティ担当者におすすめです。視聴により、自己ホスト型AI環境のアーキテクチャと運用方法、段階的な導入アプローチについて理解が深まります。
📄 このページの紹介文は AI が独自に生成したものであり、著作権をはじめとする他者の権利(商標権・名誉権・プライバシー等)を侵害しないよう配慮しています。動画の著作権は各作成者に帰属します。