At a glance Problem: Adapting large language models to specialized, high-stakes domains is slow, expensive, and hard to reproduce.  What we built: AutoAdapt automates planning, strategy selection (e.g., RAG vs. fine-tuning), and tuning under real deployment constraints.  How it works:  A structured configuration graph maps  

So maybe I’ll first turn it over to Amy. Can you tell us a little bit about your job at Microsoft and what got you into this space, maybe a little bit of your story? AMY LUERS: So as you said, I lead the sustainability science and innovation in the Microsoft Corp sustainability team, which really means I get to work with really smart people around the company, around the world, at MSR [Microsoft Research], on shaping and informing sustainability […]

[MUSIC FADES]  JAIME TEEVAN: Hi, I’m Jaime Teevan, chief scientist and technical fellow at Microsoft, and today, we’re going to talk about the new future of work.  So back in 2020, researchers from across Microsoft came together to try to make sense of this seismic shift in work practices that was happening as a result of the pandemic, and the next year, the group published the very first New Future of Work report. Microsoft has been publishing a new report every year since with no shortages of disruptions and major technological shifts in between.  Joining me today to explore the latest […]

At a glance AI benchmarks report performance on specific tasks but provide limited insight into underlying capabilities; ADeLe evaluates models by scoring both tasks and models across 18 core abilities, enabling direct comparison between task demands and model capabilities. Using these ability scores, the  

[MUSIC FADES]  I’d like to ask each of my guests to introduce themselves. Tell me a little bit about your background and what you’re currently working on—to the extent you can talk about it—in AI. So, Nicolò, would you please start?  NICOLÒ FUSI: Yeah, thank you, Doug, for having us and having me here. It’s so much fun. So I’m Nicolò Fusi. I’m a researcher at MSR [Microsoft Research]. So Doug is my boss, so I will be very, very, very good to Doug in this podcast.   No, but jokes aside, my own background is in Bayesian nonparametric. That’s what I started studying. So […]

At a glance Problem: Debugging AI agent failures is hard because trajectories are long, stochastic, and often multi-agent, so the true root cause gets buried. Solution: AgentRx (opens in new tab) pinpoints the first unrecoverable (“critical failure”) step by synthesizing guarded, executable constraints  

At a glance Today’s AI agents store long interaction histories but struggle to reuse them effectively. Raw memory retrieval can overwhelm agents with lengthy, low-value context. PlugMem transforms interaction history into structured, reusable knowledge. A single, general-purpose memory module improves performance across diverse agent benchmarks while using fewer memory tokens.  

Transcript [MUSIC]  DOUG BURGER: AI is going to reshape the future. I don’t think there’s any question about that now. How we reshape it depends on the choices we make, and so it’s important to understand what we think those shapes are.  This is The Shape of Things to Come. I’m Doug Burger. I manage Microsoft Research’s worldwide labs, and I’m excited to introduce this new Microsoft Research Podcast series.   I called the podcast The Shape of Things to Come because as researchers, the problems that we choose to solve and the technologies that we develop do […]

At a glance Today’s AI agent benchmarks test one task at a time, while real workplace productivity requires managing dozens of interdependent tasks at once. To reflect this, we created a setting called Multi-Horizon Task Environments (MHTEs). Under multi-task loads, leading computer-using agents degrade sharply, with completion rates dropping from 16.7% to 8.7%. CORPGEN introduces digital employees,