<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Rohan Kartik. Notes.</title><description>Principal Product Designer. Notes on craft, hiring, design systems, and the seams between roles.</description><link>https://rohankartik.com/</link><language>en</language><item><title>The cost of fast decisions</title><link>https://rohankartik.com/notes/the-cost-of-fast-decisions/</link><guid isPermaLink="true">https://rohankartik.com/notes/the-cost-of-fast-decisions/</guid><description>On the trade-off between velocity and the kind of decision that needs to be slept on.</description><pubDate>Sun, 12 Apr 2026 00:00:00 GMT</pubDate></item><item><title>What I look for in a design portfolio</title><link>https://rohankartik.com/notes/what-i-look-for-in-a-design-portfolio/</link><guid isPermaLink="true">https://rohankartik.com/notes/what-i-look-for-in-a-design-portfolio/</guid><description>Not the work itself. The captions, the gaps, and what gets left out.</description><pubDate>Mon, 02 Mar 2026 00:00:00 GMT</pubDate></item><item><title>The model is the only reviewer who reads every line of the file</title><link>https://rohankartik.com/notes/the-model-reads-every-line/</link><guid isPermaLink="true">https://rohankartik.com/notes/the-model-reads-every-line/</guid><description>AI in design is less about generation and more about review. The model is the third reviewer who reads every line without fatigue.</description><pubDate>Tue, 17 Feb 2026 00:00:00 GMT</pubDate></item><item><title>Decisions with no data and no time</title><link>https://rohankartik.com/notes/decisions-no-data-no-time/</link><guid isPermaLink="true">https://rohankartik.com/notes/decisions-no-data-no-time/</guid><description>The principal-level decision is usually made with bad data and a deadline. The honest framework is smaller than people think.</description><pubDate>Thu, 12 Sep 2024 00:00:00 GMT</pubDate></item><item><title>A retro is a leadership decision before it is a feedback session</title><link>https://rohankartik.com/notes/retro-is-a-leadership-decision/</link><guid isPermaLink="true">https://rohankartik.com/notes/retro-is-a-leadership-decision/</guid><description>The design lead&apos;s retro entry sets the temperature for the next sprint. The shorter the entry, the more room the team has to lead.</description><pubDate>Fri, 10 Nov 2023 00:00:00 GMT</pubDate></item><item><title>When the design system stops being the answer</title><link>https://rohankartik.com/notes/when-the-system-stops-being-the-answer/</link><guid isPermaLink="true">https://rohankartik.com/notes/when-the-system-stops-being-the-answer/</guid><description>A design system solves the wrong problem once consistency is no longer the bottleneck. Most teams keep solving the wrong problem.</description><pubDate>Mon, 10 Apr 2023 00:00:00 GMT</pubDate></item><item><title>Principal is a shape of week, not a row on a ladder</title><link>https://rohankartik.com/notes/principal-is-a-week-shape/</link><guid isPermaLink="true">https://rohankartik.com/notes/principal-is-a-week-shape/</guid><description>What separates principal from staff is not scope of work but compression of context across unrelated decisions in the same day.</description><pubDate>Wed, 15 Mar 2023 00:00:00 GMT</pubDate></item><item><title>Speed versus craft is the wrong axis</title><link>https://rohankartik.com/notes/speed-craft-false-binary/</link><guid isPermaLink="true">https://rohankartik.com/notes/speed-craft-false-binary/</guid><description>Teams who argue about speed versus craft are missing the real axis. The real one is what gets revisited and what does not.</description><pubDate>Mon, 20 Feb 2023 00:00:00 GMT</pubDate></item><item><title>The error screen is the test of the brand</title><link>https://rohankartik.com/notes/error-screen-tests-the-brand/</link><guid isPermaLink="true">https://rohankartik.com/notes/error-screen-tests-the-brand/</guid><description>A brand holds at the moment of failure or it does not hold. The error screen is where you find out.</description><pubDate>Fri, 10 Feb 2023 00:00:00 GMT</pubDate></item><item><title>Working with PMs who are smarter than the design</title><link>https://rohankartik.com/notes/pms-smarter-than-the-design/</link><guid isPermaLink="true">https://rohankartik.com/notes/pms-smarter-than-the-design/</guid><description>The good PM thinks better than the design at the strategy layer. The designer&apos;s job is to think better at the artifact layer. Confusing the two ruins both.</description><pubDate>Wed, 08 Feb 2023 00:00:00 GMT</pubDate></item><item><title>The hiring rubric that survives a year</title><link>https://rohankartik.com/notes/hiring-rubric-survives-a-year/</link><guid isPermaLink="true">https://rohankartik.com/notes/hiring-rubric-survives-a-year/</guid><description>Most design rubrics fail twelve months in because they grade for the interview, not for the role. Three criteria hold up.</description><pubDate>Wed, 25 Jan 2023 00:00:00 GMT</pubDate></item><item><title>Designing without research is a discipline, not a shortcut</title><link>https://rohankartik.com/notes/designing-without-research/</link><guid isPermaLink="true">https://rohankartik.com/notes/designing-without-research/</guid><description>Good judgment under research constraints has a teachable structure. Most designers treat it as a personality trait.</description><pubDate>Sun, 20 Nov 2022 00:00:00 GMT</pubDate></item><item><title>The artefact two teams write together is the artefact they trust</title><link>https://rohankartik.com/notes/shared-artefact-builds-trust/</link><guid isPermaLink="true">https://rohankartik.com/notes/shared-artefact-builds-trust/</guid><description>Two languages and two cadences inside one company become one when both sides have to write into the same document.</description><pubDate>Sat, 12 Nov 2022 00:00:00 GMT</pubDate></item><item><title>Critique that lands</title><link>https://rohankartik.com/notes/critique-that-lands/</link><guid isPermaLink="true">https://rohankartik.com/notes/critique-that-lands/</guid><description>Most design critique fails for one of three reasons. Each has a specific fix that costs the giver thirty seconds.</description><pubDate>Tue, 20 Sep 2022 00:00:00 GMT</pubDate></item><item><title>Find the object that is the source of truth</title><link>https://rohankartik.com/notes/find-the-object/</link><guid isPermaLink="true">https://rohankartik.com/notes/find-the-object/</guid><description>If a physical or regulated object can already hold the truth, the software around it gets to be simple.</description><pubDate>Fri, 15 Nov 2019 00:00:00 GMT</pubDate></item></channel></rss>