<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Bhargav Kowshik</title><description>Senior data scientist working on noisy real-world sensor data. Moving toward neurotech and biosignal applications.</description><link>https://bkowshik.in/</link><item><title>From 86% to 71%: learning to trust a brain decoder&apos;s accuracy</title><link>https://bkowshik.in/posts/from-86-to-71-trusting-a-brain-decoder/</link><guid isPermaLink="true">https://bkowshik.in/posts/from-86-to-71-trusting-a-brain-decoder/</guid><description>I trained a tiny CNN to read MEG brain signals on Colab. The interesting part wasn&apos;t the model — it was how stratified splits, checkpoint hygiene, and cross-validation deflated a flattering 86% into an honest 71% ± 5%, and revealed exactly where the model wins.</description><pubDate>Sun, 14 Jun 2026 07:00:00 GMT</pubDate></item><item><title>How I use Google Colab</title><link>https://bkowshik.in/posts/how-i-use-google-colab/</link><guid isPermaLink="true">https://bkowshik.in/posts/how-i-use-google-colab/</guid><description>Notes and recipes from my day-to-day Google Colab workflow — setup, data handling, and package installs.</description><pubDate>Wed, 10 Jun 2026 03:30:00 GMT</pubDate></item></channel></rss>