Sign in

Advait Sarkar

@advaitsarkar.bsky.social
61 followers 55 following 13 posts

Researcher at Cambridge & UCL studying GenAI and knowledge work

PostsRepliesMedia
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
We're trying to learn from that rather than repeat it, which means shipping something real, watching how people actually use it, and expecting to be wrong about plenty. There's a lot of work still ahead!
000
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
This is a first version, not a finished thing. The history of education is littered with technologies that promised to put a great teacher in front of every child and didn't.
100
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
Our whitepaper on the learning science foundation of the Study and Learn agent: cdn-dynmedia-1.microsoft.com/is/content/m...
cdn-dynmedia-1.microsoft.com
100
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
We've also published a whitepaper on the learning science behind it (with Michael J. Jabbour and Mark Sparvell), where I was able to make the case for something I believe deeply: that AI in education should stand alongside teachers, not take their place.
100
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
I had the privilege of helping Anoo Padte shape how the product behaves, and it's some of the work I've been happiest to be involved in.
100
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
What helps is the right amount of support at the right moment, the thing a good tutor does when they let you strain and then step in with one useful question. Bringing that kind of experience within reach of every learner is the real opportunity, and that's a different question from AI or no AI.
100
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
Going from not knowing to knowing takes effort, and there's no getting around it. I'd have said the same. However, struggle on its own isn't the goal. A learner sitting alone and stuck isn't learning, they're just stuck.
100
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
It asks questions, it nudges, it makes you sit with the difficulties, because that's where the learning actually is. The obvious objection is that not using AI also makes you do the work, so maybe this is just a band-aid on a problem AI created.
100
Advait Sarkar @advaitsarkar.bsky.social · 16/07/2026
Many AI assistants race to give you the answer as fast as they can. Study and Learn, a new agent in M365 Copilot for education users, does something almost heretical: it refuses. [Thread]
110
Advait Sarkar @advaitsarkar.bsky.social · 04/03/2026
I spoke with @philippramer.at from @derstandard.at. We discussed metacognition in human-AI work, the enduring value of learning to program, vibe coding as a powerful prototyping tool, and the promise of generative UX, all crisply distilled by Philip. www.derstandard.at/story/300000...
derstandard.at
KI-Forscher: "Müssen mehr übers Denken nachdenken"
KI macht uns schneller – aber womöglich auch flacher. Advait Sarkar hat untersucht, was passiert, wenn Wissensarbeiter immer mehr ihres Denkens auslagern
060
Advait Sarkar @advaitsarkar.bsky.social · 01/12/2025
Research Internship in Human-AI interaction (2026): Improving Grounding in Language Models Please apply: apply.careers.microsoft.com/careers/job/... #research #HCI #AI #internship #opportunity #MicrosoftResearch
apply.careers.microsoft.com
Research Internship: Improving Grounding in Language Models (NLP, HCI) | Microsoft Careers
The intern will lead a literature review across linguistics and the philosophy of language to develop a taxonomy of grounding and support relations for AI‑generated statements. They will contribute de...
042
Advait Sarkar @advaitsarkar.bsky.social · 22/10/2025
I sat down (virtually) with The Recursive chief editor Ana Marija Kostanic to discuss AI as a Tool for Thought, maintaining intentionality in knowledge work, vibe coding, and the future of the programming profession. therecursive.com/tedai-vienna...
therecursive.com
TEDAI Vienna: Advait Sarkar on Designing AI That Challenges, So People Think (Better)
“I’m here today to talk about thinking for yourself,” opened Advait Sarkar, Senior Researcher at Microsoft Research. What ensued was one of the most
010
Reposted by Advait Sarkar
Wey Wey Web @weyweyweb.bsky.social · 15/10/2025
Why visit a website when AI can do it for you? @advaitsarkar.bsky.social explores how AI is changing the way we build and use the web. From adaptive interfaces to blurred roles, what comes next might not involve a website at all. Details: weyweyweb.com/speakers/?sp... #AI #DigitalExperience
Advait Sarkar's talk
011
Reposted by Advait Sarkar
Mattias Rost @rost.me · 17/06/2025
We’re hosting a workshop at Aarhus 2025: The End of Programming (as we know it) Rethinking coding in the age of AI—beyond productivity, toward new practices, tools & ideas. 📅 Aug 19 (hybrid) 📝 Submit by June 27 🔗 mi.sh.se/~shmnjn07/th...
mi.sh.se
You are being redirected to https://glossy-twister-5ec.notion.site/Workshop-The-End-of-Programming-1c7e06f235ea80b09717ce04663c448e
012
Advait Sarkar @advaitsarkar.bsky.social · 22/04/2025
We’re presenting several papers and holding a workshop at the CHI conference next week. This explains how it all knits together
221
Reposted by Advait Sarkar
Hao-Ping (Hank) Lee @hankhplee.bsky.social · 28/03/2025
How does using GenAI tools reshape knowledge workers’ critical thinking? Our #CHI2025 paper studied 319 knowledge workers to dive into this question. w/@advaitsarkar.bsky.social @levlevlev.bsky.social Ian, Sean, Richard, Nick @msftresearch.bsky.social @hcii.cmu.edu www.microsoft.com/en-us/resear...
1146
Reposted by Advait Sarkar
Lev Tankelevitch @levlevlev.bsky.social · 13/01/2025
📣#CHI2025 workshop: Tools for Thought: Research and Design for Understanding, Protecting, and Augmenting Human Cognition with Generative AI 🌐https://aka.ms/toolsforthought-chi2025workshop 🗓️Submission deadline: Feb 13, 2025 AoE #HCI #cogsci #AI #GenerativeAI #GenAI
Promotional flyer for the CHI'25 Workshop, "Tools for Thought: Research and Design for Understanding, Protecting, and Augmenting Human Cognition with Generative AI," happening April 26, 2025, in Yokohama, Japan. Submission deadline: February 13, 2025, AoE. Workshop website: [https://aka.ms/toolsforthought-chi2025workshop](https://aka.ms/toolsforthought-chi2025workshop). Key topics: theories and perspectives for understanding, protecting, and augmenting thinking with GenAI; challenges in designing GenAI systems; measuring GenAI's impact; its effect on human skills and workflows; and guidelines for designing protective and augmentative GenAI systems. Organizers include experts from Microsoft Research, Harvard, Carnegie Mellon, University of Chicago, Tableau, University of Toronto, IBM Research, UCL, and Stanford. Logos of these institutions and the CHI 2025 logo appear at the bottom. The flyer features an illustration of a pensive person with a crowned bird on their head and the text "TOOLS FOR THOUGHT."
052
Reposted by Advait Sarkar
Richard Banks @rbanks2.bsky.social · 19/12/2024
The Tools for Thought team, that I'm a part of (aka.ms/toolsforthou...) contributed to the USC Annenberg Relevence Report on AI. Released annually from the Center for Public Relations. A great report, with a lot of diverse writers, including Microsoft's very own Steve Clayton. Worth a look.
011