Kate Blair (she/her)
@kayellbe.bsky.social
Making disruptive tech useful @ IBM Research
Views are very much my own
Views are very much my own
Reposted by Kate Blair (she/her)
🐝 BeeAI Framework Update: introducing RequirementAgent
An experimental feature tackling a core challenge in agent development: how to keep agents autonomous AND reliable!
🧠 Agents stay flexible
🛡️ But must obey your critical rules
👉 Docs: framework.beeai.dev/experimental...
#AIagents #opensource
An experimental feature tackling a core challenge in agent development: how to keep agents autonomous AND reliable!
🧠 Agents stay flexible
🛡️ But must obey your critical rules
👉 Docs: framework.beeai.dev/experimental...
#AIagents #opensource
Requirement Agent - BeeAI Framework
framework.beeai.dev
June 3, 2025 at 6:34 PM
🐝 BeeAI Framework Update: introducing RequirementAgent
An experimental feature tackling a core challenge in agent development: how to keep agents autonomous AND reliable!
🧠 Agents stay flexible
🛡️ But must obey your critical rules
👉 Docs: framework.beeai.dev/experimental...
#AIagents #opensource
An experimental feature tackling a core challenge in agent development: how to keep agents autonomous AND reliable!
🧠 Agents stay flexible
🛡️ But must obey your critical rules
👉 Docs: framework.beeai.dev/experimental...
#AIagents #opensource
Reposted by Kate Blair (she/her)
We created ACP to break down barriers between #AI #agents- enabling open, interoperable, and framework-agnostic communication.
🧠 Thrilled to see our colleagues at @ibm-research.bsky.social share our vision for what’s ahead for agentic AI.
📖 Read the new blog here!
research.ibm.com/blog/agent-c...
🧠 Thrilled to see our colleagues at @ibm-research.bsky.social share our vision for what’s ahead for agentic AI.
📖 Read the new blog here!
research.ibm.com/blog/agent-c...
An open-source protocol for AI agents to interact
IBM’s Agent Communication Protocol gives AI agents a shared language to connect and collaborate to carry out complex, real-world tasks.
research.ibm.com
May 28, 2025 at 6:10 PM
We created ACP to break down barriers between #AI #agents- enabling open, interoperable, and framework-agnostic communication.
🧠 Thrilled to see our colleagues at @ibm-research.bsky.social share our vision for what’s ahead for agentic AI.
📖 Read the new blog here!
research.ibm.com/blog/agent-c...
🧠 Thrilled to see our colleagues at @ibm-research.bsky.social share our vision for what’s ahead for agentic AI.
📖 Read the new blog here!
research.ibm.com/blog/agent-c...
Reposted by Kate Blair (she/her)
It’s funny that you can go through 20 years of schooling without ever seeing the idea that writing is a tool for thinking.
May 20, 2025 at 2:31 PM
It’s funny that you can go through 20 years of schooling without ever seeing the idea that writing is a tool for thinking.
Reposted by Kate Blair (she/her)
✨Explore how #ACP is set to become the internet protocol for AI agents by standardizing communication and enabling collaboration across autonomous systems.✨
📖 Dive into the future of interoperable AI in our new blog and ACP quickstart: towardsdatascience.com/acp-the-inte...
#aiagents #ai #agents
📖 Dive into the future of interoperable AI in our new blog and ACP quickstart: towardsdatascience.com/acp-the-inte...
#aiagents #ai #agents
ACP: The Internet Protocol for AI Agents | Towards Data Science
ACP aims to be the "HTTP of agent communication,” transforming our current landscape of siloed agents into interoperable agentic systems.
towardsdatascience.com
May 9, 2025 at 3:17 PM
✨Explore how #ACP is set to become the internet protocol for AI agents by standardizing communication and enabling collaboration across autonomous systems.✨
📖 Dive into the future of interoperable AI in our new blog and ACP quickstart: towardsdatascience.com/acp-the-inte...
#aiagents #ai #agents
📖 Dive into the future of interoperable AI in our new blog and ACP quickstart: towardsdatascience.com/acp-the-inte...
#aiagents #ai #agents
Reposted by Kate Blair (she/her)
“And I’m like, what do you mean, you ‘do your own research’? You running a double-blind study in your living room, dawg?”
- A guy walking ahead of me with his friends on a NYC sidewalk in 2021, also my favorite overheard dialogue of the entire pandemic
- A guy walking ahead of me with his friends on a NYC sidewalk in 2021, also my favorite overheard dialogue of the entire pandemic
April 30, 2025 at 11:36 AM
“And I’m like, what do you mean, you ‘do your own research’? You running a double-blind study in your living room, dawg?”
- A guy walking ahead of me with his friends on a NYC sidewalk in 2021, also my favorite overheard dialogue of the entire pandemic
- A guy walking ahead of me with his friends on a NYC sidewalk in 2021, also my favorite overheard dialogue of the entire pandemic
BeeAI is building the Agent Communication Protocol as part of our mission as a @linuxfoundation.org project.
A priority concern defining an "AgentFile" to enable offline discovery, define consumption procedures, and streamline deployment of AI agents.
Weigh in! github.com/i-am-bee/bee...
A priority concern defining an "AgentFile" to enable offline discovery, define consumption procedures, and streamline deployment of AI agents.
Weigh in! github.com/i-am-bee/bee...
Agent Manifest File Format · i-am-bee beeai · Discussion #427
As we move from the Pre-alpha to the Alpha stage, we're revisiting the structure of our agent manifests. Goals for the new manifest format: Include all necessary data for offline discovery, includi...
github.com
March 19, 2025 at 8:56 PM
BeeAI is building the Agent Communication Protocol as part of our mission as a @linuxfoundation.org project.
A priority concern defining an "AgentFile" to enable offline discovery, define consumption procedures, and streamline deployment of AI agents.
Weigh in! github.com/i-am-bee/bee...
A priority concern defining an "AgentFile" to enable offline discovery, define consumption procedures, and streamline deployment of AI agents.
Weigh in! github.com/i-am-bee/bee...
You can get 🐝 up and running with just two lines of code!
brew install i-am-bee/beeai/beeai
brew services start beeai
brew install i-am-bee/beeai/beeai
brew services start beeai
IBM's BeeAI
The open-source platform to discover, run, and compose AI agents from any framework. BeeAI helps you discover, run, and compose AI agents across any framework or language.
The open-source platform to discover, run, and compose AI agents from any framework. BeeAI helps you discover, run, and compose AI agents across any framework or language.
March 18, 2025 at 1:43 PM
You can get 🐝 up and running with just two lines of code!
brew install i-am-bee/beeai/beeai
brew services start beeai
brew install i-am-bee/beeai/beeai
brew services start beeai
Reposted by Kate Blair (she/her)
💥 Introducing BeeAI Alpha!💥
The open source platform to discover, run, and compose AI agents from ANY framework.
Try BeeAI today to:
🌐 Run agents from any framework
🔎 Search an agent catalog to find and integrate agents
⚙️ Connect specialized agents to create workflows
& more!
Try now at beeai.dev
The open source platform to discover, run, and compose AI agents from ANY framework.
Try BeeAI today to:
🌐 Run agents from any framework
🔎 Search an agent catalog to find and integrate agents
⚙️ Connect specialized agents to create workflows
& more!
Try now at beeai.dev
March 14, 2025 at 5:59 PM
💥 Introducing BeeAI Alpha!💥
The open source platform to discover, run, and compose AI agents from ANY framework.
Try BeeAI today to:
🌐 Run agents from any framework
🔎 Search an agent catalog to find and integrate agents
⚙️ Connect specialized agents to create workflows
& more!
Try now at beeai.dev
The open source platform to discover, run, and compose AI agents from ANY framework.
Try BeeAI today to:
🌐 Run agents from any framework
🔎 Search an agent catalog to find and integrate agents
⚙️ Connect specialized agents to create workflows
& more!
Try now at beeai.dev
Reposted by Kate Blair (she/her)
🔍 LLMs struggle with tables—but how robust are they really?
💡 We introduce ToRR, a benchmark for Table Reasoning and Robustness, that goes beyond accuracy, testing real-world robustness across formats & tasks.
arxiv.org/pdf/2502.19412
More details on ToRR's unique evaluation are provided below ⬇️
💡 We introduce ToRR, a benchmark for Table Reasoning and Robustness, that goes beyond accuracy, testing real-world robustness across formats & tasks.
arxiv.org/pdf/2502.19412
More details on ToRR's unique evaluation are provided below ⬇️
arxiv.org
March 9, 2025 at 7:08 PM
🔍 LLMs struggle with tables—but how robust are they really?
💡 We introduce ToRR, a benchmark for Table Reasoning and Robustness, that goes beyond accuracy, testing real-world robustness across formats & tasks.
arxiv.org/pdf/2502.19412
More details on ToRR's unique evaluation are provided below ⬇️
💡 We introduce ToRR, a benchmark for Table Reasoning and Robustness, that goes beyond accuracy, testing real-world robustness across formats & tasks.
arxiv.org/pdf/2502.19412
More details on ToRR's unique evaluation are provided below ⬇️
The Bee Team just released an updated GUI that makes it easy for anyone to build delightful, custom agents powered by open-source models. github.com/i-am-bee/bee-stack.
November 20, 2024 at 9:14 PM
The Bee Team just released an updated GUI that makes it easy for anyone to build delightful, custom agents powered by open-source models. github.com/i-am-bee/bee-stack.