Scholé CEO Vinitra Swamy on the learning science behind adaptive AI tutors
Cohere · youtube · 2026-10-31
In a Cohere Labs talk, Scholé AI co-founder and CEO Vinitra Swamy explained how LLMs enable learning systems that model what a learner knows, decide what comes next, and generate practice and explanations in real time. She mapped decades of learning science—retrieval, mastery learning, the zone of proximal development, knowledge tracing, active learning, feedback, productive struggle—onto the architecture of a multi-agent tutoring platform built from learner models, knowledge graphs, and pedagogical agents. The talk's central design question: what should we personalize, when should the system intervene, and how do we build AI that helps people learn rather than just making answers easier to get? Swamy is the first PhD graduate of EPFL's ML for education lab and previously led ONNX engineering at Microsoft AI.
More from Apps
- YC-backed Hickory launches AI-native compliance OS for banks and fintechs — ycombinator · 2026-10-31
- ChatGPT Ads Now in 89 Countries, Adds Wider Ad Unit and Multi-Image Beta — OnlineInference · 2026-10-31
- Sentience Launches Multiplayer Personal AI in Slack — ditzikow · 2026-10-31
- 7 ChatGPT photo prompts that turn selfies into pro travel, magazine and cinematic shots — ifioknkem · 2026-10-31
- Dev builds Greia, an open-source Linux OS for toddlers with 45 learning activities, using Claude — orblabs · 2026-10-31
- Decoding Google's Knowledge Graph: 5B entities, 500B facts, 194 access values — gaganghotra_ · 2026-10-31