<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>rohitwrites</title><description>Writing on tech, software engineering, and side projects.</description><link>https://rohitbele.com/</link><item><title>The Tax You Can&apos;t Avoid Paying</title><link>https://rohitbele.com/blog/the-tax-you-cant-avoid-paying/</link><guid isPermaLink="true">https://rohitbele.com/blog/the-tax-you-cant-avoid-paying/</guid><description>A pattern I keep running into across very different parts of learning: every real shortcut turns out to just move the difficulty somewhere less visible, not remove it.</description><pubDate>Sun, 16 Aug 2026 00:00:00 GMT</pubDate></item><item><title>Information Is Not Knowledge</title><link>https://rohitbele.com/blog/information-is-not-knowledge/</link><guid isPermaLink="true">https://rohitbele.com/blog/information-is-not-knowledge/</guid><description>Being able to retrieve an answer and understanding it are different capacities, and it&apos;s never been easier to confuse them for each other.</description><pubDate>Fri, 24 Jul 2026 00:00:00 GMT</pubDate></item><item><title>Lessons From Implementing a Neural Network From Scratch</title><link>https://rohitbele.com/blog/implementing-a-neural-network-from-scratch/</link><guid isPermaLink="true">https://rohitbele.com/blog/implementing-a-neural-network-from-scratch/</guid><description>Backpropagation is simple in theory and unforgiving in practice. Here&apos;s what actually went wrong when I built one with nothing but numpy, and how gradient checking saved me.</description><pubDate>Fri, 03 Jul 2026 00:00:00 GMT</pubDate></item><item><title>Learning in Public Is Harder Than It Sounds</title><link>https://rohitbele.com/blog/learning-in-public-is-harder-than-it-sounds/</link><guid isPermaLink="true">https://rohitbele.com/blog/learning-in-public-is-harder-than-it-sounds/</guid><description>The advice is simple to state and uncomfortable to actually follow, for reasons that have less to do with time and more to do with ego.</description><pubDate>Fri, 19 Jun 2026 00:00:00 GMT</pubDate></item><item><title>LLMs and the Changing Role of Software Engineers</title><link>https://rohitbele.com/blog/llms-and-the-changing-role-of-engineers/</link><guid isPermaLink="true">https://rohitbele.com/blog/llms-and-the-changing-role-of-engineers/</guid><description>The part of the job that&apos;s shrinking isn&apos;t the part most people worry about. It&apos;s not thinking that&apos;s getting outsourced — it&apos;s typing.</description><pubDate>Thu, 28 May 2026 00:00:00 GMT</pubDate></item><item><title>How to Actually Get Through a Hard Technical Book</title><link>https://rohitbele.com/blog/how-to-study-a-hard-technical-book/</link><guid isPermaLink="true">https://rohitbele.com/blog/how-to-study-a-hard-technical-book/</guid><description>Not reading advice in general — specifically what to do with a book full of exercises and proofs that&apos;s too dense to read the way you&apos;d read anything else.</description><pubDate>Tue, 05 May 2026 00:00:00 GMT</pubDate></item><item><title>What Building Data Pipelines Taught Me About Reliability</title><link>https://rohitbele.com/blog/data-pipeline-reliability-lessons/</link><guid isPermaLink="true">https://rohitbele.com/blog/data-pipeline-reliability-lessons/</guid><description>Pipelines rarely fail loudly. They fail by quietly producing plausible-looking wrong answers, which is a much harder problem than a crash.</description><pubDate>Tue, 14 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Attention Is the Scarce Resource</title><link>https://rohitbele.com/blog/attention-is-the-scarce-resource/</link><guid isPermaLink="true">https://rohitbele.com/blog/attention-is-the-scarce-resource/</guid><description>We talk about time management, but time was never really the constraint. What actually runs out, faster and less predictably, is attention.</description><pubDate>Wed, 25 Mar 2026 00:00:00 GMT</pubDate></item><item><title>A System for Studying Things I Don&apos;t Understand Yet</title><link>https://rohitbele.com/blog/a-system-for-studying-hard-subjects/</link><guid isPermaLink="true">https://rohitbele.com/blog/a-system-for-studying-hard-subjects/</guid><description>Motivation runs out. What&apos;s replaced it is a small set of habits that make studying a hard subject less dependent on how I happen to feel that day.</description><pubDate>Fri, 06 Mar 2026 00:00:00 GMT</pubDate></item><item><title>What Gradient Descent Actually Does</title><link>https://rohitbele.com/blog/what-gradient-descent-actually-does/</link><guid isPermaLink="true">https://rohitbele.com/blog/what-gradient-descent-actually-does/</guid><description>Stripped of the vector calculus notation, gradient descent is just a rule for guessing better. Here&apos;s the mechanical version I wish someone had shown me first.</description><pubDate>Tue, 17 Feb 2026 00:00:00 GMT</pubDate></item><item><title>The Book That Changed How I Read Technical Books</title><link>https://rohitbele.com/blog/the-book-that-changed-how-i-read/</link><guid isPermaLink="true">https://rohitbele.com/blog/the-book-that-changed-how-i-read/</guid><description>Richard Hamming&apos;s The Art of Doing Science and Engineering didn&apos;t teach me a new technique. It taught me to stop reading for information and start reading for judgment.</description><pubDate>Thu, 22 Jan 2026 00:00:00 GMT</pubDate></item><item><title>Learning Machine Learning From First Principles</title><link>https://rohitbele.com/blog/learning-ml-from-first-principles/</link><guid isPermaLink="true">https://rohitbele.com/blog/learning-ml-from-first-principles/</guid><description>Notes on the gap between knowing how to call a library and actually understanding what it&apos;s doing, and why I decided to close it the slow way.</description><pubDate>Thu, 08 Jan 2026 00:00:00 GMT</pubDate></item></channel></rss>