<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
    <id>https://omarkamali.com/</id>
    <title>Omar Kamali</title>
    <updated>2026-03-23T22:14:56.706Z</updated>
    <generator>awesome</generator>
    <author>
        <name>Omar Kamali</name>
        <email>omar@monitoro.co</email>
        <uri>https://omarkamali.com/</uri>
    </author>
    <link rel="alternate" href="https://omarkamali.com/"/>
    <link rel="self" href="https://omarkamali.com/atom"/>
    <subtitle>I write about technical concepts in human terms.</subtitle>
    <logo>https://omarkamali.com/assets/img/ok-social.png</logo>
    <icon>https://omarkamali.com/favicon/favicon-32x32.png</icon>
    <rights>All rights reserved 2026 &amp;copy; Omar Kamali</rights>
    <category term="Technology"/>
    <entry>
        <title type="html"><![CDATA[Beyond Tokenization: The Four Taxes and the Path Forward]]></title>
        <id>beyond-tokenization-four-taxes-path-forward</id>
        <link href="https://omarkamali.com/blog/beyond-tokenization-four-taxes-path-forward"/>
        <updated>2026-03-23T22:14:56.706Z</updated>
        <summary type="html"><![CDATA[The compounding tax stack low-resource languages carry, why vision encoders might hold the key, and the open research questions.]]></summary>
        <content type="html"><![CDATA[The compounding tax stack low-resource languages carry, why vision encoders might hold the key, and the open research questions.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[The Hidden Tax Your LLM Pays for Bad Tokenization]]></title>
        <id>hidden-tax-llm-pays-bad-tokenization</id>
        <link href="https://omarkamali.com/blog/hidden-tax-llm-pays-bad-tokenization"/>
        <updated>2026-03-23T22:14:56.566Z</updated>
        <summary type="html"><![CDATA[How bad tokenization forces language models to waste capacity on reconstruction instead of reasoning.]]></summary>
        <content type="html"><![CDATA[How bad tokenization forces language models to waste capacity on reconstruction instead of reasoning.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Tokenization is Killing Our Multilingual LLM Dream]]></title>
        <id>tokenization</id>
        <link href="https://omarkamali.com/blog/tokenization"/>
        <updated>2026-03-23T22:14:56.418Z</updated>
        <summary type="html"><![CDATA[Why tokenization is the hidden bottleneck blocking truly multilingual AI — lessons from building Sawalni and Wikilangs.]]></summary>
        <content type="html"><![CDATA[Why tokenization is the hidden bottleneck blocking truly multilingual AI — lessons from building Sawalni and Wikilangs.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Why I stopped trusting the official Wikipedia dataset, and what I did about it]]></title>
        <id>wikipedia-monthly-pipeline</id>
        <link href="https://omarkamali.com/blog/wikipedia-monthly-pipeline"/>
        <updated>2026-03-07T20:58:14.790Z</updated>
        <summary type="html"><![CDATA[It all started with a DM from a friend, member and contributor to the Moroccan Wikipedia community.
"Are you using the current version of Wikipedia? The official dataset is severely outdated. We added so many cool articles nowhere on huggingface" He was right. I was running a 2023 snapshot in 2025.]]></summary>
        <content type="html"><![CDATA[It all started with a DM from a friend, member and contributor to the Moroccan Wikipedia community.
"Are you using the current version of Wikipedia? The official dataset is severely outdated. We added so many cool articles nowhere on huggingface" He was right. I was running a 2023 snapshot in 2025.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[A Wordle for the Worldle]]></title>
        <id>worldle-for-the-world-wikilangs-launch</id>
        <link href="https://omarkamali.com/blog/worldle-for-the-world-wikilangs-launch"/>
        <updated>2026-03-04T01:09:53.142Z</updated>
        <summary type="html"><![CDATA[I built a word game for more than 300 languages, each drawing on its own Wikipedia as the source. Here's the thing nobody tells you: building a simple word game for most of these languages meant building things that didn't exist.]]></summary>
        <content type="html"><![CDATA[I built a word game for more than 300 languages, each drawing on its own Wikipedia as the source. Here's the thing nobody tells you: building a simple word game for most of these languages meant building things that didn't exist.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Picomon 0.2.0: From AMD Crash Fix to GPU Monitoring That Doesn’t Suck]]></title>
        <id>picomon-amd-nvidia-apple-silicon-gpu-monitoring</id>
        <link href="https://omarkamali.com/blog/picomon-amd-nvidia-apple-silicon-gpu-monitoring"/>
        <updated>2025-12-27T18:14:25.908Z</updated>
        <summary type="html"><![CDATA[Earlier this month, I whipped up a Python script with an LLM that parsed amd-smi output. It was ugly. It worked. I called it picomon.]]></summary>
        <content type="html"><![CDATA[Earlier this month, I whipped up a Python script with an LLM that parsed amd-smi output. It was ugly. It worked. I called it picomon.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Introducing Wikipedia Monthly: Fresh, Clean Wikipedia Dumps for NLP & AI Research]]></title>
        <id>wikipedia-monthly-fresh-clean-dumps-nlp-ai-research</id>
        <link href="https://omarkamali.com/blog/wikipedia-monthly-fresh-clean-dumps-nlp-ai-research"/>
        <updated>2025-07-19T22:36:45.738Z</updated>
        <summary type="html"><![CDATA[Announcing Wikipedia Monthly, an always fresh dataset to support research for low-resource languages]]></summary>
        <content type="html"><![CDATA[Announcing Wikipedia Monthly, an always fresh dataset to support research for low-resource languages]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Getting Perfectly Structured Data from LLMs]]></title>
        <id>how-to-get-perfectly-structured-data-from-llms</id>
        <link href="https://omarkamali.com/blog/how-to-get-perfectly-structured-data-from-llms"/>
        <updated>2025-02-23T22:44:56.022Z</updated>
        <summary type="html"><![CDATA[If you've ever struggled to get consistent JSON output from large language models, I have a simple and clever solution for you.]]></summary>
        <content type="html"><![CDATA[If you've ever struggled to get consistent JSON output from large language models, I have a simple and clever solution for you.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[2024: A Year of Growth, Innovation, and Community]]></title>
        <id>2024-a-year-of-growth-innovation-community</id>
        <link href="https://omarkamali.com/blog/2024-a-year-of-growth-innovation-community"/>
        <updated>2025-01-04T01:50:31.474Z</updated>
        <summary type="html"><![CDATA[As we leave 2024 behind, I found myself reflecting over the holidays on a transformative year that reshaped my grasp of technology's role in human connection.]]></summary>
        <content type="html"><![CDATA[As we leave 2024 behind, I found myself reflecting over the holidays on a transformative year that reshaped my grasp of technology's role in human connection.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Datapluck: Portability Tool for Huggingface Datasets]]></title>
        <id>datapluck</id>
        <link href="https://omarkamali.com/blog/datapluck"/>
        <updated>2024-09-05T02:03:25.632Z</updated>
        <summary type="html"><![CDATA[Exporting & importing Hugging Face datasets to spreadsheets and various file formats.]]></summary>
        <content type="html"><![CDATA[Exporting & importing Hugging Face datasets to spreadsheets and various file formats.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[OpenAI RAG: A Step Forward or a Step Too Far?]]></title>
        <id>openai-rag</id>
        <link href="https://omarkamali.com/blog/openai-rag"/>
        <updated>2023-11-10T23:31:16.821Z</updated>
        <summary type="html"><![CDATA[OpenAI's data integrations in Assistant and GPTs is causing ripples in the AI world. Beyond the excitement, let's look at OpenAI's strategy critically, the tension in its ecosystem, questionable comparisons to Apple, and the impending threat of commodification that OpenAI itself may face.]]></summary>
        <content type="html"><![CDATA[OpenAI's data integrations in Assistant and GPTs is causing ripples in the AI world. Beyond the excitement, let's look at OpenAI's strategy critically, the tension in its ecosystem, questionable comparisons to Apple, and the impending threat of commodification that OpenAI itself may face.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Web Scraping: Data For Everyone]]></title>
        <id>web-scraping-data-for-everyone</id>
        <link href="https://omarkamali.com/blog/web-scraping-data-for-everyone"/>
        <updated>2023-10-30T00:56:54.621Z</updated>
        <summary type="html"><![CDATA[We know what Data is, but where does it come from? With Web Scraping you can collect data from any website. This article will get you started in this world.]]></summary>
        <content type="html"><![CDATA[We know what Data is, but where does it come from? With Web Scraping you can collect data from any website. This article will get you started in this world.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[What is Data: A Brief Guide]]></title>
        <id>what-is-data-a-brief-guide</id>
        <link href="https://omarkamali.com/blog/what-is-data-a-brief-guide"/>
        <updated>2023-10-25T15:29:24.938Z</updated>
        <summary type="html"><![CDATA[An introduction to ease you into the world of data, what it is, what is it useful for and privacy concerns, as a preamble to the Data Series.]]></summary>
        <content type="html"><![CDATA[An introduction to ease you into the world of data, what it is, what is it useful for and privacy concerns, as a preamble to the Data Series.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Shaping the Future with Sawalni: The Dawn of Moroccan AI]]></title>
        <id>shaping-the-future-with-sawalni-dawn-of-moroccan-ai</id>
        <link href="https://omarkamali.com/blog/shaping-the-future-with-sawalni-dawn-of-moroccan-ai"/>
        <updated>2023-10-17T18:48:24.897Z</updated>
        <summary type="html"><![CDATA[I've been asked multiple times, "Why are you creating a Moroccan AI?" Today I want to share the story behind Sawalni, the first AI in history to speak our beautiful Moroccan Darija, with all of you.]]></summary>
        <content type="html"><![CDATA[I've been asked multiple times, "Why are you creating a Moroccan AI?" Today I want to share the story behind Sawalni, the first AI in history to speak our beautiful Moroccan Darija, with all of you.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Tokens, the secret language of AI]]></title>
        <id>tokens-the-secret-language-of-ai</id>
        <link href="https://omarkamali.com/blog/tokens-the-secret-language-of-ai"/>
        <updated>2023-06-26T02:06:39.137Z</updated>
        <summary type="html"><![CDATA[Have you ever wondered how AI systems make sense of the vast amount of information they encounter? Let's look at AI tokens and why you should care.]]></summary>
        <content type="html"><![CDATA[Have you ever wondered how AI systems make sense of the vast amount of information they encounter? Let's look at AI tokens and why you should care.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[DNS for non-Techies: Domains, Website Performance, and Content restrictions]]></title>
        <id>dns-for-non-techies-domains-website-performance-and-content-restrictions</id>
        <link href="https://omarkamali.com/blog/dns-for-non-techies-domains-website-performance-and-content-restrictions"/>
        <updated>2023-06-14T20:25:11.480Z</updated>
        <summary type="html"><![CDATA[New websites are launched every day and others stop existing overnight. Servers and IP addresses get repurposed. How does DNS keep it all together?]]></summary>
        <content type="html"><![CDATA[New websites are launched every day and others stop existing overnight. Servers and IP addresses get repurposed. How does DNS keep it all together?]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[How I built my newsletter using Gmail and Google Sheets]]></title>
        <id>a-newsletter-using-gmail-google-sheets</id>
        <link href="https://omarkamali.com/blog/a-newsletter-using-gmail-google-sheets"/>
        <updated>2023-03-12T00:06:08.035Z</updated>
        <summary type="html"><![CDATA[I've recently wanted to add a newsletter to my blog. I decided to build it instead of using a marketing tool.]]></summary>
        <content type="html"><![CDATA[I've recently wanted to add a newsletter to my blog. I decided to build it instead of using a marketing tool.]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[It starts with a message: Internet, DNS & Networks for non-Techies]]></title>
        <id>it-starts-with-a-message-internet-dns-networks-for-non-techies-part-1</id>
        <link href="https://omarkamali.com/blog/it-starts-with-a-message-internet-dns-networks-for-non-techies-part-1"/>
        <updated>2023-03-10T23:38:50.936Z</updated>
        <summary type="html"><![CDATA[There are over 22 billion machines connected to the internet. How do these machines communicate with each other to fulfil their purpose?]]></summary>
        <content type="html"><![CDATA[There are over 22 billion machines connected to the internet. How do these machines communicate with each other to fulfil their purpose?]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[What happens when you open a TikTok?]]></title>
        <id>what-happens-when-you-open-a-tiktok</id>
        <link href="https://omarkamali.com/blog/what-happens-when-you-open-a-tiktok"/>
        <updated>2023-03-09T20:40:37.000Z</updated>
        <summary type="html"><![CDATA[Do you ever stop and think about the countless invisible components working behind the scenes to bring an app like TikTok to life on your phone?]]></summary>
        <content type="html"><![CDATA[Do you ever stop and think about the countless invisible components working behind the scenes to bring an app like TikTok to life on your phone?]]></content>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[How the Internet Works: The Essential Guide to Digital Infrastructure]]></title>
        <id>essential-guide-to-digital-infrastructure-how-the-internet-works</id>
        <link href="https://omarkamali.com/blog/essential-guide-to-digital-infrastructure-how-the-internet-works"/>
        <updated>2023-03-06T20:29:59.000Z</updated>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[The Market Decides Your Project Scope]]></title>
        <id>the-market-decides-your-project-scope</id>
        <link href="https://omarkamali.com/blog/the-market-decides-your-project-scope"/>
        <updated>2019-02-09T01:00:19.000Z</updated>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[Should you automate this task?]]></title>
        <id>should-you-automate-this-task</id>
        <link href="https://omarkamali.com/blog/should-you-automate-this-task"/>
        <updated>2019-02-05T01:27:44.000Z</updated>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
    <entry>
        <title type="html"><![CDATA[First Post]]></title>
        <id>first-post</id>
        <link href="https://omarkamali.com/blog/first-post"/>
        <updated>2019-02-03T01:06:07.000Z</updated>
        <author>
            <name>Omar Kamali</name>
            <email>omar@monitoro.co</email>
            <uri>https://omarkamali.com/</uri>
        </author>
    </entry>
</feed>