<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: cahaya</title><link>https://news.ycombinator.com/user?id=cahaya</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 29 May 2026 20:30:13 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=cahaya" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by cahaya in "Ask HN: What Are You Working On? (March 2026)"]]></title><description><![CDATA[
<p>Building Cursor IDE, but for knowledge workers. Domain will go live soon on <a href="https://document.bot" rel="nofollow">https://document.bot</a></p>
]]></description><pubDate>Mon, 09 Mar 2026 10:53:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=47307372</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=47307372</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47307372</guid></item><item><title><![CDATA[New comment by cahaya in "Claude Code Remote Control"]]></title><description><![CDATA[
<p>Hoping OpenAI/ Codex will launch this soon too.</p>
]]></description><pubDate>Wed, 25 Feb 2026 11:27:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=47150150</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=47150150</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47150150</guid></item><item><title><![CDATA[New comment by cahaya in "Show HN: A real-time strategy game that AI agents can play"]]></title><description><![CDATA[
<p>Nice. Curious about 5.3-codex-high results</p>
]]></description><pubDate>Wed, 25 Feb 2026 11:00:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=47149964</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=47149964</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47149964</guid></item><item><title><![CDATA[New comment by cahaya in "Anthropic officially bans using subscription auth for third party use"]]></title><description><![CDATA[
<p><p><pre><code>  Codex app-server is the interface Codex uses to power rich clients (for example, the Codex VS Code extension). Use it when you want a deep integration inside your own product.
</code></pre>
It mentions 'Inside your own product', but not sure if that means also your own commercial application.</p>
]]></description><pubDate>Thu, 19 Feb 2026 10:12:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=47072159</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=47072159</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47072159</guid></item><item><title><![CDATA[New comment by cahaya in "Anthropic officially bans using subscription auth for third party use"]]></title><description><![CDATA[
<p>Same question here. A while ago I read rumors OpenAI might build a "Login with OpenAI" (comparable to login with Apple, Facebook, Google) so people can also use their existing sub in commercial apps. Hope it's true.</p>
]]></description><pubDate>Thu, 19 Feb 2026 10:08:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47072127</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=47072127</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47072127</guid></item><item><title><![CDATA[New comment by cahaya in "Immersa: Open-source Web-based 3D Presentation Tool"]]></title><description><![CDATA[
<p>looks nice!</p>
]]></description><pubDate>Sat, 20 Dec 2025 14:10:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=46336325</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=46336325</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46336325</guid></item><item><title><![CDATA[New comment by cahaya in "DeepSeek OCR"]]></title><description><![CDATA[
<p>Lot's of OCR/ LLM's (Even Gemini Pro 2.5) still struggle converting complex tables to markdown or HTML:  Tables with multiple headers and merged cells that get mixed up, multiple columns with tick boxes get mixed up, multi page tables that are not understood correctly. Also Llamaindex fails miserably on those things.<p>Curious to hear which OCR/ LLM excels with these specific issues? Example complex table:
<a href="https://cdn.aviation.bot/complex-tables.zip" rel="nofollow">https://cdn.aviation.bot/complex-tables.zip</a><p>I can only parse this table correctly by first parsing the table headers manually into HTML as example output. However, it still mixes up tick boxes. Full table examples:
<a href="https://www.easa.europa.eu/en/icao-compliance-checklist" rel="nofollow">https://www.easa.europa.eu/en/icao-compliance-checklist</a></p>
]]></description><pubDate>Mon, 20 Oct 2025 09:27:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=45641806</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=45641806</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45641806</guid></item><item><title><![CDATA[New comment by cahaya in "Show HN: Run Qwen3-Next-80B on 8GB GPU at 1tok/2s throughput"]]></title><description><![CDATA[
<p>Got 32GB so was hoping I could use ollm to offload it to my SSD. Slower but making it possible to run bigger models (in emergencies)</p>
]]></description><pubDate>Tue, 23 Sep 2025 16:57:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=45349735</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=45349735</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45349735</guid></item><item><title><![CDATA[New comment by cahaya in "Show HN: Run Qwen3-Next-80B on 8GB GPU at 1tok/2s throughput"]]></title><description><![CDATA[
<p>Nice. Seems like i cannot run this on my Apple silicon M chips right?</p>
]]></description><pubDate>Tue, 23 Sep 2025 13:29:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=45346766</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=45346766</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45346766</guid></item><item><title><![CDATA[New comment by cahaya in "Show HN: Pgmcp, an MCP server to query any Postgres database in natural language"]]></title><description><![CDATA[
<p>Nice, but can somebody tell me if this performs better than my simple Postgres MCP using npx? My current setup uses the LLM to search through my local Postgres in multiple steps. I guess this Pgmcp is doing multiple steps in the background and returns the final result to the LLM calling the MCP tool?<p>Codex:
```
[mcp_servers.postgresMCP]
command = "npx"
args = ["-y", "@modelcontextprotocol/server-postgres", "postgresql://user:password@localhost:5432/db"]
```<p>Cursor:
```
"postgresMCP": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-postgres",
        "postgresql://user:password@localhost:5432/db"
      ]
    },
```<p>With my setup i can easily switch between LLM's</p>
]]></description><pubDate>Thu, 18 Sep 2025 12:00:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=45288554</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=45288554</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45288554</guid></item><item><title><![CDATA[New comment by cahaya in "I got the highest score on ARC-AGI again swapping Python for English"]]></title><description><![CDATA[
<p>Are there any existing scripts/ tools to use these evolutionary algorithms also at home with e.g. Codex/GPT-5 / Claude Code?</p>
]]></description><pubDate>Wed, 17 Sep 2025 12:44:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=45275186</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=45275186</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45275186</guid></item><item><title><![CDATA[New comment by cahaya in "How does air pollution impact your brain?"]]></title><description><![CDATA[
<p>Is it me or is the link only opening as JSON?<p>{
"@context": [
"<a href="https://www.w3.org/ns/activitystreams" rel="nofollow">https://www.w3.org/ns/activitystreams</a>",
{
"Hashtag": "as:Hashtag",
"sensitive": "as:sensitive",
"dcterms": "<a href="http://purl.org/dc/terms/" rel="nofollow">http://purl.org/dc/terms/</a>"
}
],
"id": "<a href="https://neurofrontiers.blog/?p=11319" rel="nofollow">https://neurofrontiers.blog/?p=11319</a>",
"type": "Note",
"attachment": [
{
"type": "Image",
"url": "<a href="https://i0.wp.com/neurofrontiers.blog/wp-content/uploads/2025/07/brain-traffic-jam-1.jpg?fit=1024%2C576&#038;ssl=1" rel="nofollow">https://i0.wp.com/neurofrontiers.blog/wp-content/uploads/202...</a>",
"mediaType": "image/jpeg",
"name": "A cartoon depicting a traffic jam of identical black cars, each driven by a brain. Thick clouds of exhaust rise and gather above the stalled vehicles."
}
],
"attributedTo": "<a href="https://neurofrontiers.blog/author/neuronerdb/" rel="nofollow">https://neurofrontiers.blog/author/neuronerdb/</a>",
"audience": "<a href="https://neurofrontiers.blog/?author=0" rel="nofollow">https://neurofrontiers.blog/?author=0</a>",
"content": "<h2>How does air pollution</p>
]]></description><pubDate>Mon, 15 Sep 2025 13:42:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=45249664</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=45249664</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45249664</guid></item><item><title><![CDATA[New comment by cahaya in "ChatGPT Developer Mode: Full MCP client access"]]></title><description><![CDATA[
<p>I tried adding Context7 Documentation MCP and got this<p>URL:<a href="https://mcp.context7.com/mcp" rel="nofollow">https://mcp.context7.com/mcp</a>
Safety Scan: Passed<p>This MCP server can't be used by ChatGPT to search information because it doesn't implement our specification: search action not found
<a href="https://platform.openai.com/docs/mcp#create-an-mcp-server" rel="nofollow">https://platform.openai.com/docs/mcp#create-an-mcp-server</a></p>
]]></description><pubDate>Thu, 11 Sep 2025 10:47:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=45210025</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=45210025</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45210025</guid></item><item><title><![CDATA[New comment by cahaya in "Show HN: Project management system for Claude Code"]]></title><description><![CDATA[
<p>I was also looking for a video. The concept sounds good, but feels like I need to learn a lot of new commands, or have a cheat sheet next to me to be able to be able to use the framework.</p>
]]></description><pubDate>Wed, 20 Aug 2025 12:22:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=44961242</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=44961242</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44961242</guid></item><item><title><![CDATA[New comment by cahaya in "NautilusTrader: Open-source algorithmic trading platform"]]></title><description><![CDATA[
<p>Wondering how average users can benefit from this platform with Claude Code and the relation to Vending Bench that tracks how much money LLM's can make.<p><a href="https://andonlabs.com/evals/vending-bench">https://andonlabs.com/evals/vending-bench</a></p>
]]></description><pubDate>Wed, 06 Aug 2025 14:35:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=44812603</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=44812603</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44812603</guid></item><item><title><![CDATA[New comment by cahaya in "Up to date prices for LLM APIs all in one place"]]></title><description><![CDATA[
<p>Nice! Missing a cost calculator with input and output fields.</p>
]]></description><pubDate>Fri, 25 Jul 2025 13:40:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=44683009</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=44683009</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44683009</guid></item><item><title><![CDATA[New comment by cahaya in "Show HN: Conductor, a Mac app that lets you run a bunch of Claude Codes at once"]]></title><description><![CDATA[
<p>Exactly. If Conductor would work with local branches I would switch from Crystal.</p>
]]></description><pubDate>Mon, 21 Jul 2025 11:19:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=44633871</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=44633871</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44633871</guid></item><item><title><![CDATA[New comment by cahaya in "LibreOffice slams Microsoft for locking in Office users w/ complex file formats"]]></title><description><![CDATA[
<p>I can confirm. When trying convert simple Word sentences and tables to e.g. Markdown/HTML from a Word XML you need a PhD in XML edge cases and nested garbage.</p>
]]></description><pubDate>Fri, 18 Jul 2025 19:14:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=44608679</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=44608679</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44608679</guid></item><item><title><![CDATA[New comment by cahaya in "Ask HN: What are you working on? (April 2025)"]]></title><description><![CDATA[
<p>Building a AI-assisted aviation regulation compliance tool for aviation professionals: <a href="https://aviation.bot" rel="nofollow">https://aviation.bot</a>
At the moment just a RAG with EU aviation regulations. Soon FAA, UK CAA, and more complex AI agent features that do more complex deep research.</p>
]]></description><pubDate>Mon, 28 Apr 2025 09:12:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=43819176</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=43819176</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43819176</guid></item><item><title><![CDATA[New comment by cahaya in "Cohere Launches Embed 4"]]></title><description><![CDATA[
<p>Wondering how this compares to the Gemini (preview) embeddings as they seem to perform significantly better than OpenAI embeddings 3 large. I don't see any MTEB scores so hard to compare.</p>
]]></description><pubDate>Tue, 15 Apr 2025 16:37:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=43695277</link><dc:creator>cahaya</dc:creator><comments>https://news.ycombinator.com/item?id=43695277</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43695277</guid></item></channel></rss>