@adamfourney.bsky.social
I am extremely excited about this work! It opens a new and useful way of using agents (addressing a new category of "FOMO" tasks). And, it's highly generalizable: monitor prices, paper pubs, GitHub, Discord -- even traffic cameras, all with a simple text prompt
SentinelStep enables AI agents to handle monitoring tasks that run for hours or days, like watching for emails or tracking prices. It works by managing when agents should check and their context, avoiding wasted resources and missed updates. msft.it/6018sC85Y
October 21, 2025 at 4:57 PM
I am extremely excited about this work! It opens a new and useful way of using agents (addressing a new category of "FOMO" tasks). And, it's highly generalizable: monitor prices, paper pubs, GitHub, Discord -- even traffic cameras, all with a simple text prompt
Extremely excited to share this work with everyone -- we've been working on this since even before the Magentic One release last Fall. I'm also very glad to bring the Human back into the loop.
Magentic-UI, new from Microsoft Research, is an open-source research prototype of a human-centered AI agent, designed to work with people to complete complex, web-based tasks in real time over a web browser. msft.it/6013SjYgN
May 19, 2025 at 5:15 PM
Extremely excited to share this work with everyone -- we've been working on this since even before the Magentic One release last Fall. I'm also very glad to bring the Human back into the loop.
Reposted
I am presenting tomorrow (Wednesday) my TOCHI Microsoft work “Generation Probabilities Are Not Enough: Uncertainty Highlighting in AI Code Completions” at ~11:55am in G301 :) should be fun 🥳
work done w/ @jennwv.bsky.social @qveraliao.bsky.social @adamfourney.bsky.social @gaganbansal.bsky.social 🤩
work done w/ @jennwv.bsky.social @qveraliao.bsky.social @adamfourney.bsky.social @gaganbansal.bsky.social 🤩
4. On Wednesday, @helenasresearch.bsky.social will present the TOCHI paper she led on the potential of uncertainty highlighting for fostering appropriate reliance on AI-powered code completion tools.
Program link: programs.sigchi.org/chi/2025/pro...
Program link: programs.sigchi.org/chi/2025/pro...
April 29, 2025 at 2:56 AM
I am presenting tomorrow (Wednesday) my TOCHI Microsoft work “Generation Probabilities Are Not Enough: Uncertainty Highlighting in AI Code Completions” at ~11:55am in G301 :) should be fun 🥳
work done w/ @jennwv.bsky.social @qveraliao.bsky.social @adamfourney.bsky.social @gaganbansal.bsky.social 🤩
work done w/ @jennwv.bsky.social @qveraliao.bsky.social @adamfourney.bsky.social @gaganbansal.bsky.social 🤩
We’ve been working on this for quite a while. Super excited to share!
Announcing AutoGen 0.4, fully reimagined library for building advanced agentic AI systems, developed to improve code quality and robustness. Its asynchronous, event-driven architecture is designed to support dynamic, scalable workflows. Learn more: www.microsoft.com/en-us/resear...
AutoGen v0.4: Advancing the development of agentic AI systems
Announcing AutoGen 0.4, fully reimagined library for building advanced agentic AI systems, developed to improve code quality and robustness. Its asynchronous, event-driven architecture is designed to ...
www.microsoft.com
January 14, 2025 at 6:02 PM
We’ve been working on this for quite a while. Super excited to share!
Reposted
☀️New paper!
Generative AI agents are powerful but complex—how do we design them for transparency and human control? 🤖✨
At the heart of this challenge is establishing common ground, a concept from human communication. We identify 12 key challenges in improving common ground between humans & agents.
Generative AI agents are powerful but complex—how do we design them for transparency and human control? 🤖✨
At the heart of this challenge is establishing common ground, a concept from human communication. We identify 12 key challenges in improving common ground between humans & agents.
December 3, 2024 at 6:19 PM
☀️New paper!
Generative AI agents are powerful but complex—how do we design them for transparency and human control? 🤖✨
At the heart of this challenge is establishing common ground, a concept from human communication. We identify 12 key challenges in improving common ground between humans & agents.
Generative AI agents are powerful but complex—how do we design them for transparency and human control? 🤖✨
At the heart of this challenge is establishing common ground, a concept from human communication. We identify 12 key challenges in improving common ground between humans & agents.