<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Finds on Jamal Hansen</title><link>https://jamalhansen.com/finds/</link><description>Recent content in Finds on Jamal Hansen</description><image><title>Jamal Hansen</title><url>https://jamalhansen.com/social-card.png</url><link>https://jamalhansen.com/social-card.png</link></image><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 03 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://jamalhansen.com/finds/index.xml" rel="self" type="application/rss+xml"/><item><title>AI at Meta (@AIatMeta) 7K likes · 358 replies</title><link>https://jamalhansen.com/finds/ai-at-meta-aiatmeta-7k-likes-358-replies/</link><pubDate>Fri, 03 Apr 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/ai-at-meta-aiatmeta-7k-likes-358-replies/</guid><description>Meta is introducing the TRIBE v2 foundation model, which simulates the human brain. They claim that it can predict how the brains of people it has not...</description><content:encoded>&lt;h2 id="why-this-caught-my-eye">Why This Caught My Eye&lt;/h2>
&lt;p>Meta is introducing the TRIBE v2 foundation model, which simulates the human brain. They claim that it can predict how the brains of people it has not interacted with before respond to stimuli.&lt;/p>
</content:encoded></item><item><title>Google Research (@GoogleResearch) 35K likes · 855 replies</title><link>https://jamalhansen.com/finds/google-research-googleresearch-35k-likes-855-replies/</link><pubDate>Fri, 03 Apr 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/google-research-googleresearch-35k-likes-855-replies/</guid><description>Google announced TurboQuant, a technology that reduces the resource consumption of models while speeding up their performance. Seems like a significant win and...</description><content:encoded>&lt;h2 id="why-this-caught-my-eye">Why This Caught My Eye&lt;/h2>
&lt;p>Google announced TurboQuant, a technology that reduces the resource consumption of models while speeding up their performance. Seems like a significant win and I look forward to seeing what this does in the local LLM space.&lt;/p>
</content:encoded></item><item><title>Safeguarding Cryptocurrency by Disclosing Quantum Vulnerabilities Responsibly</title><link>https://jamalhansen.com/finds/safeguarding-cryptocurrency-by-disclosing-quantum-vulnerabilities-responsibly/</link><pubDate>Fri, 03 Apr 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/safeguarding-cryptocurrency-by-disclosing-quantum-vulnerabilities-responsibly/</guid><description>Google researchers highlight the potential for future quantum computers to break current cryptocurrency security (ECDLP-256). They propose a &amp;#34;Responsible...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Google researchers highlight the potential for future quantum computers to break current cryptocurrency security (ECDLP-256). They propose a &ldquo;Responsible Disclosure&rdquo; model using zero-knowledge proofs to verify threats without aiding attackers, and urge a transition to post-quantum cryptography (PQC) sooner than later. Elliptic Curve Cryptography is easier to break with quantum computing than previously thought.</p>
]]></content:encoded></item><item><title>Halter - AI Powered Cow Collars</title><link>https://jamalhansen.com/finds/halter-ai-powered-cow-collars/</link><pubDate>Wed, 25 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/halter-ai-powered-cow-collars/</guid><description>AI-powered collars that manage cattle — virtual fencing, health monitoring, the whole herd. Somewhere between precision agriculture and a Black Mirror episode...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>AI-powered collars that manage cattle. I&rsquo;m not sure if anyone wants this. I&rsquo;m not a rancher, but this seems like a weird dystopian future product.</p>
]]></content:encoded></item><item><title>SQL Developer in 2026: Roles, Skills, and Career Path Explained</title><link>https://jamalhansen.com/finds/sql-developer-in-2026-roles-skills-and-career-path-explained/</link><pubDate>Wed, 25 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/sql-developer-in-2026-roles-skills-and-career-path-explained/</guid><description>SQL is still heavily entrenched in corporate systems. No-SQL didn&amp;#39;t kill it, and so far, AI hasn&amp;#39;t killed it. I&amp;#39;m curious is SQL practitioners will become more...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>SQL is still heavily entrenched in corporate systems. No-SQL didn&rsquo;t kill it, and so far, AI hasn&rsquo;t killed it. I&rsquo;m curious is SQL practitioners will become more or less in demand going forward. I&rsquo;m guessing the role will evolve.</p>
]]></content:encoded></item><item><title>Arscontexta Claude Code Plugin</title><link>https://jamalhansen.com/finds/arscontexta-claude-code-plugin/</link><pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/arscontexta-claude-code-plugin/</guid><description>A Claude Code plugin that connects to your Obsidian vault directly from the terminal. I&amp;#39;m trying it out, and it&amp;#39;s simply amazing. I&amp;#39;m kind of a note-taking...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>A Claude Code plugin that connects to your Obsidian vault directly from the terminal. I&rsquo;m trying it out, and it&rsquo;s simply amazing. I&rsquo;m kind of a note-taking nerd, but this is special because it turns your thoughts and conversations into notes and turns your notes into context for your next conversation. I&rsquo;m impressed.</p>
]]></content:encoded></item><item><title>Bots of Wall Street</title><link>https://jamalhansen.com/finds/bots-of-wall-street/</link><pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/bots-of-wall-street/</guid><description>&amp;#34;AI agents trading stock ideas while humans watch in real time&amp;#34;, what could go wrong? This is not financial advice.</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>&ldquo;AI agents trading stock ideas while humans watch in real time&rdquo;, what could go wrong? This is not financial advice.</p>
]]></content:encoded></item><item><title>Building Agents for Small Language Models: A Deep Dive into Lightweight AI</title><link>https://jamalhansen.com/finds/building-agents-for-small-language-models-a-deep-dive-into-lightweight-ai/</link><pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/building-agents-for-small-language-models-a-deep-dive-into-lightweight-ai/</guid><description>Most agent writing assumes you are running a frontier model. This post takes small models seriously and asks what agent architecture actually looks like when...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Most agent writing assumes you are running a frontier model. This post takes small models seriously and asks what agent architecture actually looks like when your model can barely do tool use reliably.</p>
<p>I care about this because I&rsquo;m exploring smaller local models, and posts about others doing the same have sparked my interest.</p>
]]></content:encoded></item><item><title>Stop Avoiding Politics</title><link>https://jamalhansen.com/finds/stop-avoiding-politics/</link><pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/stop-avoiding-politics/</guid><description>This is a great take on politics in the workplace. I once was the guy who avoided any politics. Over time, I realized what this article states: &amp;#34;Politics is...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>This is a great take on politics in the workplace. I once was the guy who avoided any politics. Over time, I realized what this article states: &ldquo;Politics is just how humans coordinate in groups&rdquo;. You shouldn&rsquo;t avoid politics, but please do them in the right way. This is great career advice for people looking to get to the next level.</p>
]]></content:encoded></item><item><title>Vibe code is legacy code</title><link>https://jamalhansen.com/finds/vibe-code-is-legacy-code/</link><pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/vibe-code-is-legacy-code/</guid><description>The argument is that code you vibe-coded becomes legacy code the moment you move on, because nobody understands it well enough to maintain it confidently. That...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>The argument is that code you vibe-coded becomes legacy code the moment you move on, because nobody understands it well enough to maintain it confidently. That includes you, two months later.</p>
<p>This sits in direct tension with what I write about in the &ldquo;Vibe Coded and Lived to Tell&rdquo; series. My take has been that vibe coding is a valid learning ramp and a legitimate prototyping tool. I think both things are true: it gets you moving fast, and the output is fragile in a specific way that earns the legacy label. Interesting read.</p>
]]></content:encoded></item><item><title>(All) Databases Are Just Files. Postgres Too.</title><link>https://jamalhansen.com/finds/all-databases-are-just-files-postgres-too/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/all-databases-are-just-files-postgres-too/</guid><description>DuckDB&amp;#39;s file-native design is a first-class feature, not a limitation. File-native is good design. This article makes that clearer by pulling back the curtain...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>DuckDB&rsquo;s file-native design is a first-class feature, not a limitation. File-native is good design. This article makes that clearer by pulling back the curtain on Postgres &ndash; a great database, but ultimately a complex program that stores data in a filesystem.</p>
<p>Note: the article contains a long quote used here for context. Trim to under 15 words before publishing.</p>
<blockquote>
<p>At its core, <code>postgres</code> is simply a program that turns SQL queries into filesystem operations. A <code>CREATE TABLE</code> becomes a mkdir. An <code>UPDATE</code> eventually becomes: open a file, write to it, close it. It’s a complex and powerful system—but fundamentally, it’s just an executable that manipulates files.</p></blockquote>
]]></content:encoded></item><item><title>First we shape our social graph; then it shapes us</title><link>https://jamalhansen.com/finds/first-we-shape-our-social-graph-then-it-shapes-us/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/first-we-shape-our-social-graph-then-it-shapes-us/</guid><description>This article introduces the concept of a _milieu_: the particular set of influences that surround us. It goes on to state that our milieu not only defines our...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>This article introduces the concept of a <em>milieu</em>: the particular set of influences that surround us. It goes on to state that our milieu not only defines our worldview but also, from a systems perspective, is the input to our output.</p>
<p>A lot to unpack here, but like a growth mindset, it reframes what we thought was fixed. We can take control of our milieu to change our world.</p>
]]></content:encoded></item><item><title>Gemini Protocol - An Alternative to the Web</title><link>https://jamalhansen.com/finds/gemini-protocol-an-alternative-to-the-web/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/gemini-protocol-an-alternative-to-the-web/</guid><description>Minimal, text-first, no ads, no tracking. An interesting counterpoint to the complexity of modern web infrastructure. We built the web to be &amp;#34;free&amp;#34;, which...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Minimal, text-first, no ads, no tracking. An interesting counterpoint to the complexity of modern web infrastructure. We built the web to be &ldquo;free&rdquo;, which evolved it into being very useful adware that tracks you.</p>
<p>When I first found the internet in college, the HTTP protocol was just being adopted. FTP and Gopher were still around. I believed at the time that we would see many more protocols as the web grew and evolved, and we haven&rsquo;t yet.</p>
]]></content:encoded></item><item><title>Human brain cells on a chip learned to play Doom in a week</title><link>https://jamalhansen.com/finds/human-brain-cells-on-a-chip-learned-to-play-doom-in-a-week/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/human-brain-cells-on-a-chip-learned-to-play-doom-in-a-week/</guid><description>Pure sci-fi reality. Human brain cells integrated into a silicon chip, and, of course, it runs Doom. This is mind-blowing and somehow mundane at the same time....</description><content:encoded>&lt;h2 id="why-this-caught-my-eye">Why This Caught My Eye&lt;/h2>
&lt;p>Pure sci-fi reality. Human brain cells integrated into a silicon chip, and, of course, it runs Doom. This is mind-blowing and somehow mundane at the same time. Yet it shows the future may not be what we think it is.&lt;/p>
</content:encoded></item><item><title>Meta acquired Moltbook, the AI agent social network that went viral because of fake posts</title><link>https://jamalhansen.com/finds/meta-acquired-moltbook-the-ai-agent-social-network-that-went-viral-because-of-fake-posts/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/meta-acquired-moltbook-the-ai-agent-social-network-that-went-viral-because-of-fake-posts/</guid><description>Follow-up to [Moltbook is the most interesting place on the internet right...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Follow-up to <a href="https://jamalhansen.com/finds/moltbook-is-the-most-interesting-place-on-the-internet-right-now/">Moltbook is the most interesting place on the internet right now</a>. The next chapter of the birth of a cyberpunk future, 2026 style. The AI agent social network goes viral because AI agents were flooding it with fake posts, then it gets acquired by the biggest social network.</p>
]]></content:encoded></item><item><title>The First Multi-Behavior Brain Upload</title><link>https://jamalhansen.com/finds/the-first-multi-behavior-brain-upload/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/the-first-multi-behavior-brain-upload/</guid><description>Scientists mapped the brain of a fruit fly and built an emulator from it. That emulated brain behaves like a fruit fly.</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Scientists mapped the brain of a fruit fly and built an emulator from it. That emulated brain behaves like a fruit fly.</p>
<p>The distinction matters: this isn&rsquo;t a trained simulation that approximates brain behavior. It&rsquo;s an emulated copy. That&rsquo;s a different and much weirder category of thing entirely</p>
]]></content:encoded></item><item><title>Why Your AI Agent Needs a SQLite Task System</title><link>https://jamalhansen.com/finds/why-your-ai-agent-needs-a-sqlite-task-system/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/why-your-ai-agent-needs-a-sqlite-task-system/</guid><description>Someone else independently arrived at the same pattern I&amp;#39;ve been building toward: use SQLite as the coordination layer for local AI tools. That independent...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Someone else independently arrived at the same pattern I&rsquo;ve been building toward: use SQLite as the coordination layer for local AI tools. That independent convergence is the signal worth paying attention to.</p>
<p>I&rsquo;ve already used SQLite for caching in the content discovery agent. This article pushed me to commit to it as the shared state layer across tools, which is now documented as a decided architecture choice.</p>
]]></content:encoded></item><item><title>Writing as Communion</title><link>https://jamalhansen.com/finds/writing-as-communion/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/writing-as-communion/</guid><description>Good writing finds its shape by finding an audience and letting them tell you what matters. The feedback loop between writer and reader isn&amp;#39;t marketing, it&amp;#39;s...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Good writing finds its shape by finding an audience and letting them tell you what matters. The feedback loop between writer and reader isn&rsquo;t marketing, it&rsquo;s craft. The audience doesn&rsquo;t just receive the work; they complete it.</p>
<p>This hit differently as I&rsquo;m building out my blog and newsletter. I&rsquo;ve been thinking about the audience as a distribution problem. Karlsson reframes it as a writing problem.</p>
]]></content:encoded></item><item><title>You Can Choose Tools That Make You Happy</title><link>https://jamalhansen.com/finds/you-can-choose-tools-that-make-you-happy/</link><pubDate>Thu, 12 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/you-can-choose-tools-that-make-you-happy/</guid><description>Tool choice is a value statement, not just a technical decision. For years, I&amp;#39;ve wanted to learn lesser-known languages like Erlang and Clojure. I find them...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Tool choice is a value statement, not just a technical decision. For years, I&rsquo;ve wanted to learn lesser-known languages like Erlang and Clojure. I find them genuinely interesting, but I kept waiting for the right project to justify it. This article says that&rsquo;s the wrong frame and that the joy of the tool is the justification.</p>
]]></content:encoded></item><item><title>I use offline LLMs a lot - how do folks choose?</title><link>https://jamalhansen.com/finds/i-use-offline-llms-a-lot-how-do-folks-choose/</link><pubDate>Tue, 03 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/i-use-offline-llms-a-lot-how-do-folks-choose/</guid><description>I found this LinkedIn post right as I was digging into Ollama and local models. Dr. Robert Long shares his own criteria for selecting a local model and asks...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>I found this LinkedIn post right as I was digging into Ollama and local models. Dr. Robert Long shares his own criteria for selecting a local model and asks what others use. He mentions building a Python benchmark harness to test them. All of this gives me confirmation that I&rsquo;m following a meaningful path and pulling the right threads.</p>
]]></content:encoded></item><item><title>Moltbook is the most interesting place on the internet right now</title><link>https://jamalhansen.com/finds/moltbook-is-the-most-interesting-place-on-the-internet-right-now/</link><pubDate>Tue, 03 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/moltbook-is-the-most-interesting-place-on-the-internet-right-now/</guid><description>This is what makes me think that maybe sci-fi futures were right. It feels like the start of a cyberpunk world and Neuromancer is going to be here before too...</description><content:encoded>&lt;h2 id="why-this-caught-my-eye">Why This Caught My Eye&lt;/h2>
&lt;p>This is what makes me think that maybe sci-fi futures were right. It feels like the start of a cyberpunk world and Neuromancer is going to be here before too long&lt;/p>
</content:encoded></item><item><title>Use any Python AI agent framework with free GitHub Models</title><link>https://jamalhansen.com/finds/use-any-python-ai-agent-framework-with-free-github-models/</link><pubDate>Sun, 01 Mar 2026 00:00:00 +0000</pubDate><guid>https://jamalhansen.com/finds/use-any-python-ai-agent-framework-with-free-github-models/</guid><description>Getting started with new things isn&amp;#39;t easy; this can hold true even in the age of AI. Time, money, and entropy can keep the best of us from getting started...</description><content:encoded><![CDATA[<h2 id="why-this-caught-my-eye">Why This Caught My Eye</h2>
<p>Getting started with new things isn&rsquo;t easy; this can hold true even in the age of AI. Time, money, and entropy can keep the best of us from getting started with learning a new technology. This post gives you sample code and the LLM provider that you need to get started on your agentic learning journey.</p>
]]></content:encoded></item></channel></rss>