Notion Launches Developer Platform and Workers Environment, Fully Transforming into an AI Agent Orchestration Center

Notion officially launched a new developer platform, transforming its workspace into an "Artificial Intelligence Agent Hub" by introducing an AI agent orchestration layer. After the launch in February of this year, which has already created over a million "custom agents," Notion deepened its upgrades at the May 13th launch event to address limitations of existing agents, such as the inability to connect to external data and lack of custom logic.

The core of this technological advancement is the introduction of a cloud-based sandbox environment for custom code execution called "Workers," allowing developers to deploy logic code via the Notion CLI to extract and synchronize data from any external database with an API interface, such as Salesforce or Zendesk. It also supports triggering multi-step automated workflows through Webhooks. In addition, Notion enhanced ecosystem compatibility by introducing the Model Context Protocol (MCP) standard, enabling users to directly schedule third-party top-tier AI agents like Claude Code, Cursor, and Decagon within their workspaces.

Notion Launches Major Upgrade! Turn Your Workspace into a Hub for AI Assistants

Notion launches a new developer platform, officially entering the era of AI intelligence. CEO Zhao Jun stated at the launch event that Notion will transition from a collaboration note-taking tool to a central hub connecting AI agents, external data sources, and custom code, helping teams build automated workflows. Previously, Notion introduced custom agent features in February of this year, allowing users to create AI assistants.

Microsoft's MD System Outperforms GPT-5.5 in Vulnerability Detection, Amazing Capabilities!

On May 13, Microsoft launched the MDASH multi-model agent scanning framework, using a multi-agent collaboration strategy that integrates over 100 specialized AI agents based on different large or lightweight models. These agents handle code preparation and vulnerability scanning to enhance code security detection accuracy and efficiency, revolutionizing traditional single-AI model design.....

Tencent Releases OpenSearch-VL: A Comprehensive Solution for Open-Source Multimodal Deep Search Agent

Tencent Hunyuan, in collaboration with UCLA and CUHK, has open-sourced a multimodal search agent to address the evolution of Multimodal Large Language Models (MLLMs) from passive understanding to active reasoning. Previously, the lack of high-quality data, automated trajectory synthesis paths, and training recipes hindered the reproduction of top-tier agents. This open-source initiative aims to break the deadlock and advance community development....

Amazon SageMaker Introduces AI Agents to Support Model Development Driven by Natural Language

Amazon introduces AI agent capabilities in SageMaker, lowering barriers for developers to customize language models. Developers describe use cases in natural language, and the system automates model development, including training strategy recommendations, data preparation, task scheduling, and result delivery, eliminating manual API calls or data format conversions, thus reshaping development efficiency.....

Jurylu Announces Deep Collaboration with Volcano Engine, AI Short Plays Enter the Industrialization Era

Hangzhou Julilu Technology partners with Volcano Engine to integrate the Doubao video generation model Seedance 2.0, shifting AI drama production from manual workflows to industrialized processes. The core breakthrough lies in dual improvements in efficiency and quality, achieving qualitative leaps in key filmmaking metrics through the integration of Volcano Engine models and cloud infrastructure.....