<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet href="/rss/styles.xsl" type="text/xsl"?><rss version="2.0"><channel><title>📰 Read &amp; 🔄 Repeat ⏩ 💡 - Room for Ideas &amp; Inspiration</title><description>Read &amp; Repeat is my personal notebook. It&apos;s where I collect ideas and concepts that inspired me, when I read 📰 about them. And I repeat 🔄 them here to help me remember all the things I&apos;ve learned ⏩ 💡.</description><link>https://www.readrepeat.net/</link><item><title>Understanding the basic concepts &amp; building blocks of modern LLMs</title><link>https://www.readrepeat.net/inspiration/technology-behind-llm-explained-ars-technica/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/technology-behind-llm-explained-ars-technica/</guid><description>About words, tokens, vectors, context and layers. An easy-to-understand overview of how LLMs &quot;learn&quot; and how all these building blocks are joined together.
</description><pubDate>Mon, 06 May 2024 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>Don&apos;t just look for the biggest model, instead look for a right-sized model that solves your problem</title><link>https://www.readrepeat.net/inspiration/magic-llm-size-metrics-small-smart/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/magic-llm-size-metrics-small-smart/</guid><description>Is a bigger machine learning model always a better model? It might look like it, but it&apos;s not that easy. It mainly depends on what you want to achieve. Smaller can be smarter. And cheaper.
</description><pubDate>Mon, 15 Jan 2024 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>Moving From Net Promoter Score to Earned Growth Rate?!</title><link>https://www.readrepeat.net/inspiration/from-nps-to-earned-growth-rate/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/from-nps-to-earned-growth-rate/</guid><description>Working in the field of customer advocacy, I already know about the value of recommendations. Learning about a more formal approach to measure the tangible impact of customer advocacy on business results is definitely something useful to know.
</description><pubDate>Mon, 29 May 2023 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>AI is no Magic Wand for Data-Driven Business Decisions</title><link>https://www.readrepeat.net/inspiration/ai-in-business-asking-right-questions/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/ai-in-business-asking-right-questions/</guid><description>Data analytics can provide great insights and improve how a business is being run. But it&apos;s important to understand the side effects. Otherwise, you could end up optimising one metric by a little, while actually increasing overall costs. 🤷‍♂️
</description><pubDate>Mon, 22 May 2023 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>Personal Learning for Knowledge Workers with Productive Reasoning</title><link>https://www.readrepeat.net/inspiration/learning-with-productive-reasoning/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/learning-with-productive-reasoning/</guid><description>Knowing things is great. Assuming things can become an issue. Learning about how to externalise and challenge assumptions within teams has shown to be a very useful tool to improve internal communication and decision-making.
</description><pubDate>Mon, 15 May 2023 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>Microsoft, an Impressive Example of Business Reinvention</title><link>https://www.readrepeat.net/inspiration/microsoft-impressive-example-buisness-reinvention/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/microsoft-impressive-example-buisness-reinvention/</guid><description>It&apos;s really interesting how a company that almost everyone knows transformed itself - at least partially - by adopting a more modern and agile mindset which is facilitating a cultural shift from knowing everything better to active and continuous learning.
</description><pubDate>Mon, 08 May 2023 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>Doing Things Just for Fun is Good</title><link>https://www.readrepeat.net/inspiration/doing-things-just-for-fun-is-good/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/doing-things-just-for-fun-is-good/</guid><description>Many things in our lives can look like a waste of timeor appear boring to others. But many of exactly these things can also bring a smile to our face. 🙂
</description><pubDate>Thu, 04 May 2023 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>A Good Definition of Digital Transformation</title><link>https://www.readrepeat.net/inspiration/defining-digital-transformation/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/defining-digital-transformation/</guid><description>The term &quot;digital transformation&quot; is used all the time. But what does it actually mean? I heard a helpful definition in a podcast.
</description><pubDate>Mon, 24 Apr 2023 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>The Power of Writing Simply</title><link>https://www.readrepeat.net/inspiration/power-of-writing-simply-haruki-murakami/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/power-of-writing-simply-haruki-murakami/</guid><description>Compact is good. Shorter is better. Leave out, what isn&apos;t really needed. Easy? Try it!
</description><pubDate>Mon, 17 Apr 2023 08:00:00 GMT</pubDate><language>en-gb</language></item><item><title>The Concept of Operational Transparency</title><link>https://www.readrepeat.net/inspiration/the-concept-of-operational-transparency/</link><guid isPermaLink="true">https://www.readrepeat.net/inspiration/the-concept-of-operational-transparency/</guid><description>Opening up the black box by letting users see what is happening in the background, in real-time, instead of just presenting results, to increase perceived value and tolerated waiting times.
</description><pubDate>Tue, 11 Apr 2023 08:00:00 GMT</pubDate><language>en-gb</language></item></channel></rss>