<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Kaashyap Murali</title><description>Tech Lead at Microsoft working on digital agent co-workers for Copilot Studio: orchestration, evals, and memory systems.</description><link>https://kaashyapmurali.com/</link><item><title>DisasterLens: predicting building damage from satellite images</title><link>https://kaashyapmurali.com/projects/disaster-lens/</link><guid isPermaLink="true">https://kaashyapmurali.com/projects/disaster-lens/</guid><description>Per-building damage assessment from pre- and post-disaster satellite images on xBD, with Mask R-CNN building detection and a fine-tuned, satellite-pretrained DINOv3 ViT-L.</description><pubDate>Mon, 26 Oct 2026 09:00:00 GMT</pubDate></item><item><title>(Published on Microsoft&apos;s blog) Memory in Copilot Studio</title><link>https://microsoft.github.io/mcscatblog/posts/agent-memory/</link><guid isPermaLink="true">https://microsoft.github.io/mcscatblog/posts/agent-memory/</guid><description>How agent memory bridges conversations in Copilot Studio, from short- and long-term memory to reflection and evaluation.</description><pubDate>Mon, 05 Oct 2026 00:00:00 GMT</pubDate></item><item><title>DeepMind&apos;s AI Co-Mathematician and the Case for Episodic Memory</title><link>https://kaashyapmurali.com/posts/deepmind-ai-co-mathematician/</link><guid isPermaLink="true">https://kaashyapmurali.com/posts/deepmind-ai-co-mathematician/</guid><description>DeepMind&apos;s AI Co-Mathematician treats failed proof attempts as durable project state, not noise to clear. Notes on why episodic agent memory, not the next model, is the real frontier.</description><pubDate>Wed, 13 May 2026 09:00:00 GMT</pubDate></item></channel></rss>