Jonathan Abend
@jonathanabend.bsky.social
AI Agents, Agent Mesh, GraphRAG, Data Fabric.
Data & AI Manager @ Accenture
Views are my own.
Data & AI Manager @ Accenture
Views are my own.
Nice explanation of how to establish a secure identity access management for AI agents: youtu.be/DQX81oJfsTE?...
Define user/admin roles to access/develop (specific) agents.
Let the tool code handle data access. The agent only passes a session ID. The tool fetches secrets from a secret manager.
Define user/admin roles to access/develop (specific) agents.
Let the tool code handle data access. The agent only passes a session ID. The tool fetches secrets from a secret manager.
Identity and Access Management for Agents
YouTube video by Google Cloud Tech
youtu.be
November 10, 2025 at 6:21 AM
Nice explanation of how to establish a secure identity access management for AI agents: youtu.be/DQX81oJfsTE?...
Define user/admin roles to access/develop (specific) agents.
Let the tool code handle data access. The agent only passes a session ID. The tool fetches secrets from a secret manager.
Define user/admin roles to access/develop (specific) agents.
Let the tool code handle data access. The agent only passes a session ID. The tool fetches secrets from a secret manager.
One of the next things on my to do list is to try out the Amazon Q extension for VS Code. Saw it last week on the screen of my colleague.
Amazon Q - Visual Studio Marketplace share.google/yYptrSt7WhIK...
Amazon Q - Visual Studio Marketplace share.google/yYptrSt7WhIK...
Amazon Q - Visual Studio Marketplace
Extension for Visual Studio Code - The most capable generative AI–powered assistant for software development.
share.google
November 9, 2025 at 7:09 PM
One of the next things on my to do list is to try out the Amazon Q extension for VS Code. Saw it last week on the screen of my colleague.
Amazon Q - Visual Studio Marketplace share.google/yYptrSt7WhIK...
Amazon Q - Visual Studio Marketplace share.google/yYptrSt7WhIK...
Today, I tried to enhance an existing application with the Cline extension for VS Code. (It's like vibe coding with Cursor - just in VS Code). But I am not convinced of the results yet.
August 29, 2025 at 12:49 PM
Today, I tried to enhance an existing application with the Cline extension for VS Code. (It's like vibe coding with Cursor - just in VS Code). But I am not convinced of the results yet.
You probably all know how well #MCP simplifies connecting #AI to tools and other resources.
But do you also know how to integrate MCP into your enterprise architecture?
#AWS has published this nice flow diagram of how such an integration could look like:
aws.amazon.com/blogs/machin...
But do you also know how to integrate MCP into your enterprise architecture?
#AWS has published this nice flow diagram of how such an integration could look like:
aws.amazon.com/blogs/machin...
August 18, 2025 at 7:36 PM
You probably all know how well #MCP simplifies connecting #AI to tools and other resources.
But do you also know how to integrate MCP into your enterprise architecture?
#AWS has published this nice flow diagram of how such an integration could look like:
aws.amazon.com/blogs/machin...
But do you also know how to integrate MCP into your enterprise architecture?
#AWS has published this nice flow diagram of how such an integration could look like:
aws.amazon.com/blogs/machin...
Reposted by Jonathan Abend
🚀 Skyrocketing！ 🚀 (200+ new stars)
📦 OpenPipe / ART
⭐ 3,991 (+251)
🗒 Python
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, Kimi, and more!
📦 OpenPipe / ART
⭐ 3,991 (+251)
🗒 Python
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, Kimi, and more!
GitHub - OpenPipe/ART: Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, Kimi, and more!
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, Kimi, and more! - OpenP...
github.com
July 31, 2025 at 6:02 PM
Reposted by Jonathan Abend
🚀 Skyrocketing！ 🚀 (200+ new stars)
📦 unclecode / crawl4ai
⭐ 48,620 (+279)
🗒 Python
🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN
📦 unclecode / crawl4ai
⭐ 48,620 (+279)
🗒 Python
🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN
GitHub - unclecode/crawl4ai: 🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN
🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN - unclecode/crawl4ai
github.com
July 21, 2025 at 7:02 PM
Built this easy A2A example in which each AI agent microservice is both client and server, and registers itself within an agent registry (host agent).
Each agent can decompose tasks and delegate subtasks to the most suitable agent in the agent mesh.
#AI #Agentic #A2A #Mesh
github.com/jfabend/A2A_...
Each agent can decompose tasks and delegate subtasks to the most suitable agent in the agent mesh.
#AI #Agentic #A2A #Mesh
github.com/jfabend/A2A_...
GitHub - jfabend/A2A_bidirectional
Contribute to jfabend/A2A_bidirectional development by creating an account on GitHub.
github.com
July 21, 2025 at 11:12 AM
Built this easy A2A example in which each AI agent microservice is both client and server, and registers itself within an agent registry (host agent).
Each agent can decompose tasks and delegate subtasks to the most suitable agent in the agent mesh.
#AI #Agentic #A2A #Mesh
github.com/jfabend/A2A_...
Each agent can decompose tasks and delegate subtasks to the most suitable agent in the agent mesh.
#AI #Agentic #A2A #Mesh
github.com/jfabend/A2A_...