<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: cobanov</title><link>https://news.ycombinator.com/user?id=cobanov</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sun, 27 Sep 2026 02:08:54 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=cobanov" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by cobanov in "Ollaya – Ollama for open-source, Jev-style decision models"]]></title><description><![CDATA[
<p>Thanks!</p>
]]></description><pubDate>Fri, 25 Sep 2026 19:07:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=49848628</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=49848628</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49848628</guid></item><item><title><![CDATA[New comment by cobanov in "Ollaya – Ollama for open-source, Jev-style decision models"]]></title><description><![CDATA[
<p>Fair, and I'd be happy if they did. Ollaya uses the same API as Jev, so your code isn't tied to it either way</p>
]]></description><pubDate>Fri, 25 Sep 2026 19:07:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=49848625</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=49848625</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49848625</guid></item><item><title><![CDATA[New comment by cobanov in "Ollaya – Ollama for open-source, Jev-style decision models"]]></title><description><![CDATA[
<p>Depends on the model. The small ones I support today are well below Jev on harder queries, but fine for simple, well-defined questions. The open models that get close to Jev are bigger, and I'm adding support for those next.</p>
]]></description><pubDate>Fri, 25 Sep 2026 19:06:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=49848619</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=49848619</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49848619</guid></item><item><title><![CDATA[New comment by cobanov in "Ollaya – Ollama for open-source, Jev-style decision models"]]></title><description><![CDATA[
<p>The link rgbrgb posted is a good overview. The best open ones are close to Jev now, but they're big models. And I agree, if you have an eval set for a fixed task, a trained classifier is the better choice.</p>
]]></description><pubDate>Fri, 25 Sep 2026 19:06:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=49848613</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=49848613</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49848613</guid></item><item><title><![CDATA[New comment by cobanov in "Ollaya – Ollama for open-source, Jev-style decision models"]]></title><description><![CDATA[
<p>Fair point, that example is basically classification. I'll change it to something that looks more like a real decision.</p>
]]></description><pubDate>Fri, 25 Sep 2026 19:05:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=49848600</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=49848600</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49848600</guid></item><item><title><![CDATA[New comment by cobanov in "Ollaya – Ollama for open-source, Jev-style decision models"]]></title><description><![CDATA[
<p>Developer here. You're right, Laya is a lot weaker than Jev, especially on harder queries. It's a small model, so it's fast, but that's the trade-off. The open models that get close to Jev are much bigger, and running those is what I'm working on next.</p>
]]></description><pubDate>Fri, 25 Sep 2026 19:05:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=49848595</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=49848595</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49848595</guid></item><item><title><![CDATA[KV Cache and Flash Attention with interactive diagrams]]></title><description><![CDATA[
<p>Article URL: <a href="https://kvcache.cobanov.dev/">https://kvcache.cobanov.dev/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48212530">https://news.ycombinator.com/item?id=48212530</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 20 May 2026 19:08:55 +0000</pubDate><link>https://kvcache.cobanov.dev/</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=48212530</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48212530</guid></item><item><title><![CDATA[How AI agent memory works]]></title><description><![CDATA[
<p>Article URL: <a href="https://memory.cobanov.dev/">https://memory.cobanov.dev/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48040311">https://news.ycombinator.com/item?id=48040311</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 06 May 2026 19:10:04 +0000</pubDate><link>https://memory.cobanov.dev/</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=48040311</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48040311</guid></item><item><title><![CDATA[From latent spaces to JWTs: how agents taught me backend]]></title><description><![CDATA[
<p>Article URL: <a href="https://blog.cobanov.dev/blog/agents">https://blog.cobanov.dev/blog/agents</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47557518">https://news.ycombinator.com/item?id=47557518</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 28 Mar 2026 19:31:11 +0000</pubDate><link>https://blog.cobanov.dev/blog/agents</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=47557518</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47557518</guid></item><item><title><![CDATA[How I Connected My Tesla to Claude]]></title><description><![CDATA[
<p>Article URL: <a href="https://blog.cobanov.dev/blog/teslamcp">https://blog.cobanov.dev/blog/teslamcp</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=44238667">https://news.ycombinator.com/item?id=44238667</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 10 Jun 2025 16:25:42 +0000</pubDate><link>https://blog.cobanov.dev/blog/teslamcp</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=44238667</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44238667</guid></item><item><title><![CDATA[The Generative AI Revolution I Witnessed]]></title><description><![CDATA[
<p>Article URL: <a href="https://blog.cobanov.dev/blog/genai_revolution">https://blog.cobanov.dev/blog/genai_revolution</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=44227328">https://news.ycombinator.com/item?id=44227328</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 09 Jun 2025 18:09:29 +0000</pubDate><link>https://blog.cobanov.dev/blog/genai_revolution</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=44227328</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44227328</guid></item><item><title><![CDATA[Piracy]]></title><description><![CDATA[
<p>Article URL: <a href="https://blog.cobanov.cloud/blog/piracy">https://blog.cobanov.cloud/blog/piracy</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=41704649">https://news.ycombinator.com/item?id=41704649</a></p>
<p>Points: 121</p>
<p># Comments: 118</p>
]]></description><pubDate>Tue, 01 Oct 2024 04:38:56 +0000</pubDate><link>https://blog.cobanov.cloud/blog/piracy</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=41704649</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41704649</guid></item><item><title><![CDATA[Show HN: Easy Webpage Summarizer – Quickly Summarize Webpages and YouTube Videos]]></title><description><![CDATA[
<p>I'm excited to share a project. It's a Python script that utilizes the LangChain framework and the ChatOllama model to generate concise summaries from webpages and YouTube videos. For those preferring a graphical interface, it includes a Gradio app that runs in the browser to use the summarizer interactively. Easily containerize and deploy the summarizer with Docker.<p>The tool is perfect for anyone needing quick insights without reading through the entire content/ It's open for contributions, so if you're interested in improving or extending its functionalities, feel free to dive in!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40226459">https://news.ycombinator.com/item?id=40226459</a></p>
<p>Points: 11</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 01 May 2024 17:31:49 +0000</pubDate><link>https://github.com/cobanov/easy-web-summarizer</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=40226459</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40226459</guid></item><item><title><![CDATA[New comment by cobanov in "Python toolkit for image clustering using PCA and K-means with support for GPU"]]></title><description><![CDATA[
<p>Tasnif is a Python package designed for clustering images into user-defined classes based on their visual content. It utilizes deep learning to generate image embeddings, Principal Component Analysis (PCA) for dimensionality reduction, and K-means for clustering. Tasnif supports processing on both GPU and CPU, making it versatile for different computational environments.</p>
]]></description><pubDate>Tue, 12 Mar 2024 16:49:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=39681727</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=39681727</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39681727</guid></item><item><title><![CDATA[Python toolkit for image clustering using PCA and K-means with support for GPU]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/cobanov/tasnif">https://github.com/cobanov/tasnif</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39681726">https://news.ycombinator.com/item?id=39681726</a></p>
<p>Points: 1</p>
<p># Comments: 1</p>
]]></description><pubDate>Tue, 12 Mar 2024 16:49:02 +0000</pubDate><link>https://github.com/cobanov/tasnif</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=39681726</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39681726</guid></item><item><title><![CDATA[New comment by cobanov in "The Endless Acid Banger: algorithmic self-composing acid techno music"]]></title><description><![CDATA[
<p>it's sounds like discovery channel's how it's made show</p>
]]></description><pubDate>Tue, 20 Apr 2021 07:53:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=26871873</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=26871873</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=26871873</guid></item><item><title><![CDATA[New comment by cobanov in "The Endless Acid Banger: algorithmic self-composing acid techno music"]]></title><description><![CDATA[
<p>it's sounds like discovery channel's how it works show</p>
]]></description><pubDate>Tue, 20 Apr 2021 07:52:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=26871865</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=26871865</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=26871865</guid></item><item><title><![CDATA[New comment by cobanov in "Free Python course for 5 days with certificate, starts today"]]></title><description><![CDATA[
<p>If you are a beginner to Python, we will be giving a Python workshop for the next 5 days starting today. It is completely free and you will have access to free resources including quizzes and notebooks.<p>You will earn a certificate at the end of the course.<p>February 1, 2021- February 5, 2021<p>2.00pm - 4.00pm (GMT+1)<p>4.00pm - 6.00pm (GMT+3)<p>You can join in the link below:<p><a href="https://globalaihub.com/event/introduction-to-python-programming-1-5-february/" rel="nofollow">https://globalaihub.com/event/introduction-to-python-program...</a></p>
]]></description><pubDate>Mon, 01 Feb 2021 12:47:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=25988141</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=25988141</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25988141</guid></item><item><title><![CDATA[Free Python course for 5 days with certificate, starts today]]></title><description><![CDATA[
<p>Article URL: <a href="https://globalaihub.com/event/introduction-to-python-programming-1-5-february/">https://globalaihub.com/event/introduction-to-python-programming-1-5-february/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=25988140">https://news.ycombinator.com/item?id=25988140</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Mon, 01 Feb 2021 12:47:08 +0000</pubDate><link>https://globalaihub.com/event/introduction-to-python-programming-1-5-february/</link><dc:creator>cobanov</dc:creator><comments>https://news.ycombinator.com/item?id=25988140</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25988140</guid></item></channel></rss>