<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: adilkhanovkz</title><link>https://news.ycombinator.com/user?id=adilkhanovkz</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 16 Apr 2026 22:58:19 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=adilkhanovkz" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by adilkhanovkz in "The Missing Context Layer: Why Your LLM Agent Can't Do More Than Text-to-SQL"]]></title><description><![CDATA[
<p>This resonates. We see the same pattern in financial modeling - the 'context layer' for financial models doesn't exist either. Models are flat spreadsheets with no entity relationships, no dependency graphs, no semantic structure. AI can't reason about =B47*C12 any more than it can resolve a customer across Stripe and Zendesk. We're working on this problem for finance specifically - structuring models as code with typed dependencies so AI has something meaningful to work with.</p>
]]></description><pubDate>Tue, 14 Apr 2026 17:14:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=47768379</link><dc:creator>adilkhanovkz</dc:creator><comments>https://news.ycombinator.com/item?id=47768379</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47768379</guid></item><item><title><![CDATA[New comment by adilkhanovkz in "Ask HN: What Are You Working On? (April 2026)"]]></title><description><![CDATA[
<p>We're building a python library that compiles revenue = price * volume into =B4*B5 in Excel — because the final user always asks for Excel in the end. <a href="https://modeleon.ai/" rel="nofollow">https://modeleon.ai/</a></p>
]]></description><pubDate>Mon, 13 Apr 2026 13:08:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=47751433</link><dc:creator>adilkhanovkz</dc:creator><comments>https://news.ycombinator.com/item?id=47751433</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47751433</guid></item><item><title><![CDATA[New comment by adilkhanovkz in "Why AI Sucks at Front End"]]></title><description><![CDATA[
<p>I cant even actually receive good essay from it and still writing each word myself.</p>
]]></description><pubDate>Sun, 12 Apr 2026 14:40:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=47740271</link><dc:creator>adilkhanovkz</dc:creator><comments>https://news.ycombinator.com/item?id=47740271</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47740271</guid></item><item><title><![CDATA[New comment by adilkhanovkz in "Open Models have crossed a threshold"]]></title><description><![CDATA[
<p>I think open models trail frontier by maybe 6-12 months right now. But today's frontier already solves most practical problems. So give it a year — open models hit today's level, and that's good enough for most use cases. At that point the competition stops being about who trains the best model and starts being about who builds the best applications on top of models everyone has access to.</p>
]]></description><pubDate>Thu, 09 Apr 2026 19:59:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=47708968</link><dc:creator>adilkhanovkz</dc:creator><comments>https://news.ycombinator.com/item?id=47708968</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47708968</guid></item></channel></rss>