🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

You can try out proactive coding assistants built on Continue here: github.com/sebzhao/Coding… We're excited for people to use this tool for real-world studies of proactivity in developer workflows. Full demo paper: arxiv.org/abs/2503.14724

We're doing work to make Magentic-UI more extensible, MCP servers allow you to easily plug in special purpose tools to solve tasks more efficiently.

The gains (and harms) from AI depend on how we use it. Now that millions use AI daily for work, I'm surprised why companies don't invest in training people to use AI better? I see vague incentives for "using AI" but no guidance on "how". I spent 2 years in my PhD thinking about

This is really interesting! I'm really surprised by this negative result of AI for coding as it goes against some of my work and the literature so we need to understand why. I found this graph really cool as it mirrors our CUPS work on Copilot with Gagan Bansal Eric Horvitz

BREAKING NEWS! Most people aren’t prompting models with IMO problems :) They’re prompting with tasks that need more context, like “plz make talk slides.” In an ACL oral, I’ll cover challenges in human-LM grounding (in 60K+ real interactions) & introduce a benchmark: RIFTS. 🧵

Had a blast working on this during my internship at Microsoft Research with wonderful collaborators Hussein Mozannar, Gagan Bansal, Adam Fourney, and Eric Horvitz Finally, here’s the RIFTS dataset: huggingface.co/datasets/micro… And the paper: arxiv.org/abs/2503.13975

Magentic‑UI shows that putting a person back in charge of plan, pause, and approve steps lets large‑language‑model agents finish tricky web tasks more reliably. Autonomous agents today still wander, misread context, or click the wrong thing because nobody is watching in real