<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/">
    <channel>
        <title>Clint Johnson</title>
        <link>https://www.clint-johnson.com</link>
        <description>Healthcare AI consultant specializing in FHIR, EHR integrations, and clinical decision support. Built platforms processing 100K+ messages daily.</description>
        <lastBuildDate>Wed, 10 Jun 2026 02:36:34 GMT</lastBuildDate>
        <docs>https://validator.w3.org/feed/docs/rss2.html</docs>
        <generator>https://github.com/jpmonette/feed</generator>
        <image>
            <title>Clint Johnson</title>
            <url>https://www.clint-johnson.com/og-image.png</url>
            <link>https://www.clint-johnson.com</link>
        </image>
        <copyright>All rights reserved 2026</copyright>
        <item>
            <title><![CDATA[Prompt Engineering Didn't Die. It Got Unrolled.]]></title>
            <link>https://www.clint-johnson.com/articles/prompt-engineering-got-unrolled</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/prompt-engineering-got-unrolled</guid>
            <pubDate>Mon, 01 Jun 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Everyone keeps announcing the death of prompt engineering. They are describing the symptom, not the shift. The loops you used to run by hand — refine, retry, verify, learn — moved out of your head and into infrastructure. Four of them, simultaneously.]]></description>
            <content:encoded><![CDATA[Everyone keeps announcing the death of prompt engineering. They are describing the symptom, not the shift. The loops you used to run by hand — refine, retry, verify, learn — moved out of your head and into infrastructure. Four of them, simultaneously.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/prompt-engineering-got-unrolled/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Two Rhythms of B2B Tech: What Palantir Gets Right That Most Companies Get Wrong]]></title>
            <link>https://www.clint-johnson.com/articles/palantir-foundry-forward-deployed-engineering</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/palantir-foundry-forward-deployed-engineering</guid>
            <pubDate>Sat, 16 May 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Palantir built a company on the idea that software alone is not enough — you need engineers embedded with customers. That model has a name, a cost, and a hidden technical debt time bomb that most B2B companies are quietly sitting on.]]></description>
            <content:encoded><![CDATA[Palantir built a company on the idea that software alone is not enough — you need engineers embedded with customers. That model has a name, a cost, and a hidden technical debt time bomb that most B2B companies are quietly sitting on.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/palantir-foundry-forward-deployed-engineering/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Healthcare AI Will Be Won by Verticals. The Recipe Has Been Around for a Decade.]]></title>
            <link>https://www.clint-johnson.com/articles/healthcare-ai-vertical-wins</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/healthcare-ai-vertical-wins</guid>
            <pubDate>Fri, 08 May 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Most healthcare AI companies are failing for the same reason. The ones winning are all running the same playbook — one that Palantir figured out long before anyone called it AI. Forward-deployed engineer. Ontology. Integrations. Then AI tooling, and only then.]]></description>
            <content:encoded><![CDATA[Most healthcare AI companies are failing for the same reason. The ones winning are all running the same playbook — one that Palantir figured out long before anyone called it AI. Forward-deployed engineer. Ontology. Integrations. Then AI tooling, and only then.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/healthcare-ai-vertical-wins/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Penpal: Dispatch Tool Today, RPG Interface Tomorrow]]></title>
            <link>https://www.clint-johnson.com/articles/penpal-dispatch-rpg-vision</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/penpal-dispatch-rpg-vision</guid>
            <pubDate>Thu, 07 May 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[I built a tool that turns GitHub issues into pull requests using a three-agent pipeline. That's the boring part. The interesting part is what happens when you stop thinking about AI agents as productivity tools and start thinking about them as a workforce — and you build a world for them to live in.]]></description>
            <content:encoded><![CDATA[I built a tool that turns GitHub issues into pull requests using a three-agent pipeline. That's the boring part. The interesting part is what happens when you stop thinking about AI agents as productivity tools and start thinking about them as a workforce — and you build a world for them to live in.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/penpal-dispatch-rpg-vision/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Mirth Connect, what happened? Here's What Comes Next.]]></title>
            <link>https://www.clint-johnson.com/articles/future-of-mirth</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/future-of-mirth</guid>
            <pubDate>Wed, 06 May 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Mirth democratized healthcare integration. Then NextGen acquired it, and the world moved on. After 12 years of building on top of it, here's what the next generation of integration tooling looks like — and why it was inevitable.]]></description>
            <content:encoded><![CDATA[Mirth democratized healthcare integration. Then NextGen acquired it, and the world moved on. After 12 years of building on top of it, here's what the next generation of integration tooling looks like — and why it was inevitable.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/future-of-mirth/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Product Evals in Three Steps (That You'll Actually Do)]]></title>
            <link>https://www.clint-johnson.com/articles/three-step-product-evals</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/three-step-product-evals</guid>
            <pubDate>Sat, 18 Apr 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Most teams skip evals because the process feels overwhelming. Here is the three-step framework that makes eval-driven development achievable: label a small dataset, calibrate an LLM evaluator to human judgment, then iterate configs against the harness. No excuses left.]]></description>
            <content:encoded><![CDATA[Most teams skip evals because the process feels overwhelming. Here is the three-step framework that makes eval-driven development achievable: label a small dataset, calibrate an LLM evaluator to human judgment, then iterate configs against the harness. No excuses left.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/three-step-product-evals/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Table Stakes for Pragmatic Development Using LLMs]]></title>
            <link>https://www.clint-johnson.com/articles/pragmatic-llm-development</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/pragmatic-llm-development</guid>
            <pubDate>Mon, 30 Mar 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Updated for 2026: lessons from two years of using Claude Code in production. Context engineering, real eval frameworks, model economics, and agent workflows — what actually works.]]></description>
            <content:encoded><![CDATA[Updated for 2026: lessons from two years of using Claude Code in production. Context engineering, real eval frameworks, model economics, and agent workflows — what actually works.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/pragmatic-llm-development/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Staff Engineer Layoff Survival Guide: Lessons from 2008, 2020, 2023 — and Now]]></title>
            <link>https://www.clint-johnson.com/articles/staff-advice</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/staff-advice</guid>
            <pubDate>Sun, 22 Mar 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[I've survived three tech recessions. Lost my job in one, held the axe in another. Now it's 2026 and the AI boom changed the rules. Here's the updated playbook.]]></description>
            <content:encoded><![CDATA[I've survived three tech recessions. Lost my job in one, held the axe in another. Now it's 2026 and the AI boom changed the rules. Here's the updated playbook.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/staff-advice/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Principal IC Playbook Nobody Shares With You]]></title>
            <link>https://www.clint-johnson.com/articles/principal-ic-playbook</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/principal-ic-playbook</guid>
            <pubDate>Thu, 12 Mar 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Reaching principal is the first rung of a new ladder, not the last rung of the old one. Here is what nobody told me about the ownership-autonomy paradox, how leverage actually works at this level, and the charter model I wish I had from day one.]]></description>
            <content:encoded><![CDATA[Reaching principal is the first rung of a new ladder, not the last rung of the old one. Here is what nobody told me about the ownership-autonomy paradox, how leverage actually works at this level, and the charter model I wish I had from day one.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/principal-ic-playbook/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The ESM Mess: JavaScript's Module System Is Still Broken and Here's Why]]></title>
            <link>https://www.clint-johnson.com/articles/esm-nodejs-still-broken</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/esm-nodejs-still-broken</guid>
            <pubDate>Thu, 05 Mar 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[ES Modules have been the supposed future of JavaScript for nine years. Only 9-27% of the ecosystem has actually adopted them. Here's what's really going on, and how to survive until the ecosystem commits.]]></description>
            <content:encoded><![CDATA[ES Modules have been the supposed future of JavaScript for nine years. Only 9-27% of the ecosystem has actually adopted them. Here's what's really going on, and how to survive until the ecosystem commits.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/esm-nodejs-still-broken/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Three Things Exceptional Engineering Leaders Do (And the One They Stop Doing)]]></title>
            <link>https://www.clint-johnson.com/articles/three-pillars-engineering-leadership</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/three-pillars-engineering-leadership</guid>
            <pubDate>Sun, 22 Feb 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Most engineering leaders excel at one of three pillars — providing direction, removing obstacles, or foreseeing change — and quietly fail at the other two. After 12 years and more management mistakes than I care to count, here's what I've learned about building strength across all three.]]></description>
            <content:encoded><![CDATA[Most engineering leaders excel at one of three pillars — providing direction, removing obstacles, or foreseeing change — and quietly fail at the other two. After 12 years and more management mistakes than I care to count, here's what I've learned about building strength across all three.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/three-pillars-engineering-leadership/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Open-Weight LLM Landscape in 2026: What Engineers Actually Need to Know]]></title>
            <link>https://www.clint-johnson.com/articles/open-weight-llm-trends</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/open-weight-llm-trends</guid>
            <pubDate>Sun, 15 Feb 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[The open-weight ecosystem has matured faster than most engineers realize. MoE proliferation, hybrid attention, and extended context windows are changing what's actually deployable on-premise — and that matters more than ever for healthcare AI.]]></description>
            <content:encoded><![CDATA[The open-weight ecosystem has matured faster than most engineers realize. MoE proliferation, hybrid attention, and extended context windows are changing what's actually deployable on-premise — and that matters more than ever for healthcare AI.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/open-weight-llm-trends/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Software 2.0 Is Here and It Changed How I Think About Programming]]></title>
            <link>https://www.clint-johnson.com/articles/software-2-0</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/software-2-0</guid>
            <pubDate>Wed, 11 Feb 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[In 2017, a post called "Software 2.0" argued that neural networks would replace explicit logic as the dominant programming paradigm. Nine years later, that prediction has fully landed — and the implications for how we build software are bigger than most engineers want to admit.]]></description>
            <content:encoded><![CDATA[In 2017, a post called "Software 2.0" argued that neural networks would replace explicit logic as the dominant programming paradigm. Nine years later, that prediction has fully landed — and the implications for how we build software are bigger than most engineers want to admit.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/software-2-0/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Three Ways to Know If Your Career Is Actually Growing]]></title>
            <link>https://www.clint-johnson.com/articles/three-metrics-career-growth</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/three-metrics-career-growth</guid>
            <pubDate>Wed, 04 Feb 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Normal career metrics — title, pay, team size — tell you how you're doing relative to others. They don't tell you whether you're growing. Here are three that do.]]></description>
            <content:encoded><![CDATA[Normal career metrics — title, pay, team size — tell you how you're doing relative to others. They don't tell you whether you're growing. Here are three that do.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/three-metrics-career-growth/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[RAG Isn't Dead. You're Just Using It Wrong.]]></title>
            <link>https://www.clint-johnson.com/articles/rag-isnt-dead</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/rag-isnt-dead</guid>
            <pubDate>Thu, 29 Jan 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[The 'RAG is dead' narrative is wrong — but it's wrong in an interesting way. After building RAG systems in healthcare production, here's what actually kills LLM context quality and what to do about it.]]></description>
            <content:encoded><![CDATA[The 'RAG is dead' narrative is wrong — but it's wrong in an interesting way. After building RAG systems in healthcare production, here's what actually kills LLM context quality and what to do about it.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/rag-isnt-dead/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Fine-Tuning a 70B Model on a Consumer GPU: The Q-LoRA Practical Guide]]></title>
            <link>https://www.clint-johnson.com/articles/qlora-consumer-gpu-finetuning</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/qlora-consumer-gpu-finetuning</guid>
            <pubDate>Sat, 17 Jan 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[Q-LoRA + SFTTrainer + Flash Attention v2 means you can fine-tune a 70B parameter model on 24GB of VRAM. Here is what that actually looks like end-to-end, what it costs in quality, and when you should just use the API instead.]]></description>
            <content:encoded><![CDATA[Q-LoRA + SFTTrainer + Flash Attention v2 means you can fine-tune a 70B parameter model on 24GB of VRAM. Here is what that actually looks like end-to-end, what it costs in quality, and when you should just use the API instead.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/qlora-consumer-gpu-finetuning/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Time vs. Timing: The Career Framework I Wish I Had Earlier]]></title>
            <link>https://www.clint-johnson.com/articles/time-vs-timing-career</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/time-vs-timing-career</guid>
            <pubDate>Fri, 16 Jan 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[I've made bets that paid off because of timing and bets that paid off because of compounding. Confusing the two is how careers stall.]]></description>
            <content:encoded><![CDATA[I've made bets that paid off because of timing and bets that paid off because of compounding. Confusing the two is how careers stall.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/time-vs-timing-career/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The LLM Year in Review: What Actually Mattered in 2025 (And What Was Noise)]]></title>
            <link>https://www.clint-johnson.com/articles/state-of-llms-2025</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/state-of-llms-2025</guid>
            <pubDate>Thu, 08 Jan 2026 00:00:00 GMT</pubDate>
            <description><![CDATA[The prediction was: bigger models win. The reality was: DeepSeek R1 rewrote the rules in January and nothing was the same after that. Here is what 2025 actually taught us about reasoning, inference-time compute, and the changing economics of intelligence.]]></description>
            <content:encoded><![CDATA[The prediction was: bigger models win. The reality was: DeepSeek R1 rewrote the rules in January and nothing was the same after that. Here is what 2025 actually taught us about reasoning, inference-time compute, and the changing economics of intelligence.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/state-of-llms-2025/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[From Contractor to Consultant: The Mindset Shift That Changes Your Income]]></title>
            <link>https://www.clint-johnson.com/articles/contractor-to-ai-consultant</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/contractor-to-ai-consultant</guid>
            <pubDate>Mon, 29 Dec 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Contractors sell time. Consultants sell results. After 12 years in AI and ML, I learned that distinction the hard way — and the moment it clicked changed how I price, position, and pick clients.]]></description>
            <content:encoded><![CDATA[Contractors sell time. Consultants sell results. After 12 years in AI and ML, I learned that distinction the hard way — and the moment it clicked changed how I price, position, and pick clients.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/contractor-to-ai-consultant/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[What the Teams Actually Shipping Coding Agents Have Figured Out]]></title>
            <link>https://www.clint-johnson.com/articles/production-coding-agents</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/production-coding-agents</guid>
            <pubDate>Tue, 23 Dec 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Coding agents are the most economically viable AI in production today. Here are the patterns that Devin, Cline, Amp, and others converged on — and what they mean for anyone building or using agents seriously.]]></description>
            <content:encoded><![CDATA[Coding agents are the most economically viable AI in production today. Here are the patterns that Devin, Cline, Amp, and others converged on — and what they mean for anyone building or using agents seriously.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/production-coding-agents/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Stop Shipping Features: Why AI Products Need an Experiment Mindset]]></title>
            <link>https://www.clint-johnson.com/articles/experiment-velocity-beats-features</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/experiment-velocity-beats-features</guid>
            <pubDate>Thu, 11 Dec 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[After shipping 12 features in a quarter and moving zero meaningful metrics, I learned the hard way that AI products are not software projects. The roadmap is a hypothesis board, not a delivery schedule.]]></description>
            <content:encoded><![CDATA[After shipping 12 features in a quarter and moving zero meaningful metrics, I learned the hard way that AI products are not software projects. The roadmap is a hypothesis board, not a delivery schedule.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/experiment-velocity-beats-features/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Beyond Chunks: Why Faceted Context Is the Future of RAG]]></title>
            <link>https://www.clint-johnson.com/articles/faceted-context-engineering</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/faceted-context-engineering</guid>
            <pubDate>Sun, 23 Nov 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Chunk-based RAG returns results. Faceted context gives agents peripheral vision — an understanding of the information landscape that lets them navigate rather than just consume. Here is what that looks like in a domain where getting it wrong actually matters.]]></description>
            <content:encoded><![CDATA[Chunk-based RAG returns results. Faceted context gives agents peripheral vision — an understanding of the information landscape that lets them navigate rather than just consume. Here is what that looks like in a domain where getting it wrong actually matters.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/faceted-context-engineering/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Context Engineering: The Skill That Replaced Prompt Engineering]]></title>
            <link>https://www.clint-johnson.com/articles/context-engineering-the-meta-skill</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/context-engineering-the-meta-skill</guid>
            <pubDate>Tue, 04 Nov 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[After 12 years in ML and two years building production AI systems, I stopped obsessing over prompts. The engineers who ship better agents are not writing better instructions — they are designing better information spaces.]]></description>
            <content:encoded><![CDATA[After 12 years in ML and two years building production AI systems, I stopped obsessing over prompts. The engineers who ship better agents are not writing better instructions — they are designing better information spaces.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/context-engineering-the-meta-skill/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Every Service Is Going to Need an MCP Layer]]></title>
            <link>https://www.clint-johnson.com/articles/every-service-needs-an-mcp</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/every-service-needs-an-mcp</guid>
            <pubDate>Fri, 17 Oct 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[REST APIs were designed for humans calling services through UIs. AI agents are not humans. Here is what breaks when you expose your existing APIs to agents, and what the right architecture actually looks like.]]></description>
            <content:encoded><![CDATA[REST APIs were designed for humans calling services through UIs. AI agents are not humans. Here is what breaks when you expose your existing APIs to agents, and what the right architecture actually looks like.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/every-service-needs-an-mcp/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[When English Became a Programming Language]]></title>
            <link>https://www.clint-johnson.com/articles/english-as-programming-language</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/english-as-programming-language</guid>
            <pubDate>Mon, 29 Sep 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[v0 just proved that English plus AI can replace traditional web development for most apps. I've spent 12 years mastering this craft. Here's my honest take on what that means.]]></description>
            <content:encoded><![CDATA[v0 just proved that English plus AI can replace traditional web development for most apps. I've spent 12 years mastering this craft. Here's my honest take on what that means.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/english-as-programming-language/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Fine-Tuning LLMs Without the RLHF Headache: The DPO Approach]]></title>
            <link>https://www.clint-johnson.com/articles/fine-tuning-with-dpo</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/fine-tuning-with-dpo</guid>
            <pubDate>Mon, 22 Sep 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[RLHF is the right idea with the wrong implementation cost for most teams. DPO flips the math — here's how I'd use it to align a healthcare AI model on clinician feedback without burning a month on reward model engineering.]]></description>
            <content:encoded><![CDATA[RLHF is the right idea with the wrong implementation cost for most teams. DPO flips the math — here's how I'd use it to align a healthcare AI model on clinician feedback without burning a month on reward model engineering.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/fine-tuning-with-dpo/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Agency Beats Intelligence: How I Now Hire (And Evaluate Myself)]]></title>
            <link>https://www.clint-johnson.com/articles/agency-beats-intelligence</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/agency-beats-intelligence</guid>
            <pubDate>Wed, 10 Sep 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Raw intelligence is abundant and cheap. The engineers thriving in the AI era are the ones with agency — the ability to set goals, act under uncertainty, and self-correct. Here is how I changed my interview process to find them.]]></description>
            <content:encoded><![CDATA[Raw intelligence is abundant and cheap. The engineers thriving in the AI era are the ones with agency — the ability to set goals, act under uncertainty, and self-correct. Here is how I changed my interview process to find them.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/agency-beats-intelligence/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[You Are No Longer a Coder: The Shift from Execution to Direction]]></title>
            <link>https://www.clint-johnson.com/articles/programmer-shift-direction</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/programmer-shift-direction</guid>
            <pubDate>Sat, 23 Aug 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[After 12 years of building systems by hand, I stopped writing most of my own code. Here is what changed, what I delegated to AI, what I found it cannot do, and why the hardest part of the transition had nothing to do with technology.]]></description>
            <content:encoded><![CDATA[After 12 years of building systems by hand, I stopped writing most of my own code. Here is what changed, what I delegated to AI, what I found it cannot do, and why the hardest part of the transition had nothing to do with technology.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/programmer-shift-direction/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Context Rot: The Silent Performance Killer in Your LLM Application]]></title>
            <link>https://www.clint-johnson.com/articles/context-rot-llm-performance</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/context-rot-llm-performance</guid>
            <pubDate>Tue, 05 Aug 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Your LLM system works great in demos and degrades in production. The culprit is almost never the model. It's what you're feeding it. Here's how to diagnose and fix context rot before it kills your product.]]></description>
            <content:encoded><![CDATA[Your LLM system works great in demos and degrades in production. The culprit is almost never the model. It's what you're feeding it. Here's how to diagnose and fix context rot before it kills your product.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/context-rot-llm-performance/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Why Your AI Gets Smarter When You Let It Think Longer]]></title>
            <link>https://www.clint-johnson.com/articles/why-your-ai-thinks-longer</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/why-your-ai-thinks-longer</guid>
            <pubDate>Fri, 18 Jul 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Test-time compute is the most underused lever in production AI right now. Here is what chain-of-thought, best-of-N sampling, and process reward models actually mean for practitioners building real products — and when to use them vs. just grabbing a bigger model.]]></description>
            <content:encoded><![CDATA[Test-time compute is the most underused lever in production AI right now. Here is what chain-of-thought, best-of-N sampling, and process reward models actually mean for practitioners building real products — and when to use them vs. just grabbing a bigger model.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/why-your-ai-thinks-longer/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Async-First Engineering Team: What Actually Works (And What Doesn't)]]></title>
            <link>https://www.clint-johnson.com/articles/async-first-remote-engineering</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/async-first-remote-engineering</guid>
            <pubDate>Sun, 29 Jun 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[I went async-first with my remote engineering team. Productivity went up. Culture took a hit. Here's the honest accounting of what changed, what broke, and the specific practices that made it worth it.]]></description>
            <content:encoded><![CDATA[I went async-first with my remote engineering team. Productivity went up. Culture took a hit. Here's the honest accounting of what changed, what broke, and the specific practices that made it worth it.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/async-first-remote-engineering/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[What the AI Tool Ecosystem Is Actually Telling You]]></title>
            <link>https://www.clint-johnson.com/articles/ai-oss-ecosystem-lessons</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/ai-oss-ecosystem-lessons</guid>
            <pubDate>Wed, 11 Jun 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[After wasting months on AI tools that sounded great and died within a year, I started reading the ecosystem differently. Here is the framework I use now — and why tool selection in healthcare AI is a compliance problem, not just a productivity one.]]></description>
            <content:encoded><![CDATA[After wasting months on AI tools that sounded great and died within a year, I started reading the ecosystem differently. Here is the framework I use now — and why tool selection in healthcare AI is a compliance problem, not just a productivity one.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/ai-oss-ecosystem-lessons/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[What Autonomous Vehicles Taught Me About Multi-Agent AI Design]]></title>
            <link>https://www.clint-johnson.com/articles/decentralized-agent-lessons</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/decentralized-agent-lessons</guid>
            <pubDate>Sat, 24 May 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[BAIR researchers discovered that just 5% autonomous vehicle penetration can smooth all highway traffic — with no central coordination. That finding quietly reshapes how I think about building multi-agent AI systems.]]></description>
            <content:encoded><![CDATA[BAIR researchers discovered that just 5% autonomous vehicle penetration can smooth all highway traffic — with no central coordination. That finding quietly reshapes how I think about building multi-agent AI systems.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/decentralized-agent-lessons/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Death of Institutional Memory: Why I'm Building Atlas (And Why It Matters)]]></title>
            <link>https://www.clint-johnson.com/articles/death-of-institutional-memory</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/death-of-institutional-memory</guid>
            <pubDate>Tue, 06 May 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[After watching countless organizations hemorrhage critical knowledge with every resignation, I realized we need a fundamentally different approach to capturing and preserving the networks of relationships that actually run businesses.]]></description>
            <content:encoded><![CDATA[After watching countless organizations hemorrhage critical knowledge with every resignation, I realized we need a fundamentally different approach to capturing and preserving the networks of relationships that actually run businesses.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/death-of-institutional-memory/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Vision + Language: How Multimodal LLMs Actually Work (And When to Use Them)]]></title>
            <link>https://www.clint-johnson.com/articles/multimodal-llms-vision-language</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/multimodal-llms-vision-language</guid>
            <pubDate>Mon, 05 May 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Multimodal LLMs integrate vision through two fundamentally different architectures. Knowing which one you need — and why — is the decision that shapes every other technical choice in your build.]]></description>
            <content:encoded><![CDATA[Multimodal LLMs integrate vision through two fundamentally different architectures. Knowing which one you need — and why — is the decision that shapes every other technical choice in your build.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/multimodal-llms-vision-language/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Self-Healing Stack: What AI-Native Infrastructure Actually Means]]></title>
            <link>https://www.clint-johnson.com/articles/ai-cloud-self-healing</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/ai-cloud-self-healing</guid>
            <pubDate>Thu, 17 Apr 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[The AI Cloud vision — where infrastructure monitors, optimizes, and repairs itself — is compelling. Some of it exists today. Most of it doesn't yet. Here's an honest breakdown of what self-healing infrastructure looks like in practice — and what engineers should actually be doing to prepare.]]></description>
            <content:encoded><![CDATA[The AI Cloud vision — where infrastructure monitors, optimizes, and repairs itself — is compelling. Some of it exists today. Most of it doesn't yet. Here's an honest breakdown of what self-healing infrastructure looks like in practice — and what engineers should actually be doing to prepare.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/ai-cloud-self-healing/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[FHIR Meets Graph Databases: Exploring Healthcare's Natural Network Structure]]></title>
            <link>https://www.clint-johnson.com/articles/clarity-health-clair</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/clarity-health-clair</guid>
            <pubDate>Sun, 30 Mar 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[How FHIR's interconnected resources transform into powerful graph relationships. Exploring the potential of graph technologies in healthcare AI at Clarity Health Project.]]></description>
            <content:encoded><![CDATA[How FHIR's interconnected resources transform into powerful graph relationships. Exploring the potential of graph technologies in healthcare AI at Clarity Health Project.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/clarity-health-clair/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Tools I Dropped When AI Changed My Development Workflow]]></title>
            <link>https://www.clint-johnson.com/articles/tools-dropped-for-ai-workflow</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/tools-dropped-for-ai-workflow</guid>
            <pubDate>Fri, 21 Feb 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[After 12 years of accumulating dev tools, AI coding assistants forced me to rethink every layer of my stack. Here's what I dropped, what I added, and the principle behind the whole thing.]]></description>
            <content:encoded><![CDATA[After 12 years of accumulating dev tools, AI coding assistants forced me to rethink every layer of my stack. Here's what I dropped, what I added, and the principle behind the whole thing.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/tools-dropped-for-ai-workflow/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[From GPT-2 to DeepSeek: The Architectural Changes That Actually Mattered]]></title>
            <link>https://www.clint-johnson.com/articles/llm-architecture-evolution</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/llm-architecture-evolution</guid>
            <pubDate>Tue, 04 Feb 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[I've been reading ML papers for 10 years. Most don't matter. These architectural choices did. RoPE, GQA, SwiGLU — each one solved a real scaling problem. Here's what practitioners need to know when a new model claims 'better architecture.']]></description>
            <content:encoded><![CDATA[I've been reading ML papers for 10 years. Most don't matter. These architectural choices did. RoPE, GQA, SwiGLU — each one solved a real scaling problem. Here's what practitioners need to know when a new model claims 'better architecture.']]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/llm-architecture-evolution/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Building a GenAI Platform That Doesn't Collapse Under Its Own Weight]]></title>
            <link>https://www.clint-johnson.com/articles/layered-genai-platform</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/layered-genai-platform</guid>
            <pubDate>Mon, 03 Feb 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Most GenAI platforms fail not because the models are bad, but because teams build everything at once. A practitioner's guide to layered GenAI architecture — from the minimal production-ready core to healthcare-grade guardrails and beyond.]]></description>
            <content:encoded><![CDATA[Most GenAI platforms fail not because the models are bad, but because teams build everything at once. A practitioner's guide to layered GenAI architecture — from the minimal production-ready core to healthcare-grade guardrails and beyond.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/layered-genai-platform/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The GenAI Strategy Question You're Not Asking (But Should Be)]]></title>
            <link>https://www.clint-johnson.com/articles/genai-strategy-for-practitioners</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/genai-strategy-for-practitioners</guid>
            <pubDate>Thu, 16 Jan 2025 00:00:00 GMT</pubDate>
            <description><![CDATA[Everyone asks 'how should we use GenAI?' The honest answer requires a harder question first: does AI's unique capability actually create new value here, or is it just a more expensive way to do something that already worked? A practitioner's framework for getting this right — especially in healthcare.]]></description>
            <content:encoded><![CDATA[Everyone asks 'how should we use GenAI?' The honest answer requires a harder question first: does AI's unique capability actually create new value here, or is it just a more expensive way to do something that already worked? A practitioner's framework for getting this right — especially in healthcare.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/genai-strategy-for-practitioners/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Every Failed AI Product Has the Same Root Cause]]></title>
            <link>https://www.clint-johnson.com/articles/every-failed-ai-product-root-cause</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/every-failed-ai-product-root-cause</guid>
            <pubDate>Sun, 29 Dec 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[After 12 years in ML and AI, I keep seeing the same failure pattern: teams that ship fast and iterate on vibes instead of building systematic evaluation systems. Evals are not a nice-to-have — they are the core competency of any serious AI product team.]]></description>
            <content:encoded><![CDATA[After 12 years in ML and AI, I keep seeing the same failure pattern: teams that ship fast and iterate on vibes instead of building systematic evaluation systems. Evals are not a nice-to-have — they are the core competency of any serious AI product team.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/every-failed-ai-product-root-cause/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The 6 Ways I've Watched GenAI Projects Fail (And How to Avoid Them)]]></title>
            <link>https://www.clint-johnson.com/articles/six-ways-genai-projects-fail</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/six-ways-genai-projects-fail</guid>
            <pubDate>Tue, 10 Dec 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[After 12 years in ML and two years watching GenAI projects go sideways in healthcare — sometimes with real patient consequences — here are the six failure modes I see over and over again, and what to do instead.]]></description>
            <content:encoded><![CDATA[After 12 years in ML and two years watching GenAI projects go sideways in healthcare — sometimes with real patient consequences — here are the six failure modes I see over and over again, and what to do instead.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/six-ways-genai-projects-fail/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[When to Look Beyond Standard LLMs (And When to Stop Overthinking It)]]></title>
            <link>https://www.clint-johnson.com/articles/beyond-standard-llms</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/beyond-standard-llms</guid>
            <pubDate>Mon, 11 Nov 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Most teams should use a frontier API and move on. But there are specific situations — extreme latency, long-context scale, cost walls, privacy constraints — where alternative architectures actually matter. Here's the decision framework I use.]]></description>
            <content:encoded><![CDATA[Most teams should use a frontier API and move on. But there are specific situations — extreme latency, long-context scale, cost walls, privacy constraints — where alternative architectures actually matter. Here's the decision framework I use.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/beyond-standard-llms/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[When Recommendations Meet Language: The LLM-RecSys Convergence]]></title>
            <link>https://www.clint-johnson.com/articles/llm-recsys-convergence</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/llm-recsys-convergence</guid>
            <pubDate>Mon, 04 Nov 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Most AI stacks treat the recommendation engine and the language model as two separate systems that hand off to each other. A new class of hybrid models eliminates that seam — and the implications for domain-specific AI are significant.]]></description>
            <content:encoded><![CDATA[Most AI stacks treat the recommendation engine and the language model as two separate systems that hand off to each other. A new class of hybrid models eliminates that seam — and the implications for domain-specific AI are significant.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/llm-recsys-convergence/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Trading Speed for Quality: A Practical Guide to Inference-Time Scaling]]></title>
            <link>https://www.clint-johnson.com/articles/inference-time-scaling-tradeoffs</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/inference-time-scaling-tradeoffs</guid>
            <pubDate>Thu, 17 Oct 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Inference-time scaling lets you tune the latency-quality tradeoff at runtime rather than at training time. Here is a practical framework for deciding when to use Best-of-N sampling, beam search, iterative refinement, or one-shot generation — with real examples from clinical AI.]]></description>
            <content:encoded><![CDATA[Inference-time scaling lets you tune the latency-quality tradeoff at runtime rather than at training time. Here is a practical framework for deciding when to use Best-of-N sampling, beam search, iterative refinement, or one-shot generation — with real examples from clinical AI.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/inference-time-scaling-tradeoffs/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Inside the Black Box: What Mechanistic Interpretability Means for Builders]]></title>
            <link>https://www.clint-johnson.com/articles/mechanistic-interpretability-explained</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/mechanistic-interpretability-explained</guid>
            <pubDate>Sat, 28 Sep 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Healthcare AI requires explainability — "the model said so" is not a clinical rationale. Mechanistic interpretability is the research field trying to change that. Here is what it actually offers practitioners today, what the gap still is, and what you can do in the meantime.]]></description>
            <content:encoded><![CDATA[Healthcare AI requires explainability — "the model said so" is not a clinical rationale. Mechanistic interpretability is the research field trying to change that. Here is what it actually offers practitioners today, what the gap still is, and what you can do in the meantime.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/mechanistic-interpretability-explained/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[How to Actually Test If Your AI Will Say Something Dangerous]]></title>
            <link>https://www.clint-johnson.com/articles/jailbreak-evaluation-benchmarks</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/jailbreak-evaluation-benchmarks</guid>
            <pubDate>Tue, 10 Sep 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Most teams treat jailbreak testing as a vibe check. StrongREJECT achieves 0.90 Spearman correlation with human judgment — which means automated safety evaluation is real, and there is no good excuse not to build it into your pipeline.]]></description>
            <content:encoded><![CDATA[Most teams treat jailbreak testing as a vibe check. StrongREJECT achieves 0.90 Spearman correlation with human judgment — which means automated safety evaluation is real, and there is no good excuse not to build it into your pipeline.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/jailbreak-evaluation-benchmarks/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Attack Your LLM App Is Definitely Vulnerable To]]></title>
            <link>https://www.clint-johnson.com/articles/prompt-injection-llm-security</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/prompt-injection-llm-security</guid>
            <pubDate>Fri, 23 Aug 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Prompt injection is the #1 OWASP threat to LLM applications — and most teams are not taking it seriously. Here is what the attack looks like, why it is so hard to stop, and how to actually harden your system.]]></description>
            <content:encoded><![CDATA[Prompt injection is the #1 OWASP threat to LLM applications — and most teams are not taking it seriously. Here is what the attack looks like, why it is so hard to stop, and how to actually harden your system.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/prompt-injection-llm-security/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Honest Guide to LLM Evals: What Actually Works]]></title>
            <link>https://www.clint-johnson.com/articles/honest-guide-to-llm-evals</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/honest-guide-to-llm-evals</guid>
            <pubDate>Sun, 04 Aug 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Most teams skip real evals and wonder why their AI products degrade in production. Here is the framework that actually holds up — from 30-minute manual reviews to binary scoring to knowing when your eval suite is finally doing its job.]]></description>
            <content:encoded><![CDATA[Most teams skip real evals and wonder why their AI products degrade in production. Here is the framework that actually holds up — from 30-minute manual reviews to binary scoring to knowing when your eval suite is finally doing its job.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/honest-guide-to-llm-evals/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[5 Reasons to Solve for Adoption Before Building Your Digital Health Tool]]></title>
            <link>https://www.clint-johnson.com/articles/digital-health-adoption</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/digital-health-adoption</guid>
            <pubDate>Wed, 17 Jul 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[You built a great digital health product. Clinicians love the idea. But no one is buying. Why? Too many companies start with a great idea, build the tech first, and then struggle to get adoption.]]></description>
            <content:encoded><![CDATA[You built a great digital health product. Clinicians love the idea. But no one is buying. Why? Too many companies start with a great idea, build the tech first, and then struggle to get adoption.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/digital-health-adoption/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Why Your LLM Evaluator Is Lying to You]]></title>
            <link>https://www.clint-johnson.com/articles/llm-judge-wont-save-you</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/llm-judge-wont-save-you</guid>
            <pubDate>Sat, 29 Jun 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[LLM-as-judge evaluators feel like quality assurance but behave like rubber stamps. They fail hardest on the outputs that matter most — edge cases, safety-critical errors, domain-specific nuance. Here is what to do instead.]]></description>
            <content:encoded><![CDATA[LLM-as-judge evaluators feel like quality assurance but behave like rubber stamps. They fail hardest on the outputs that matter most — edge cases, safety-critical errors, domain-specific nuance. Here is what to do instead.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/llm-judge-wont-save-you/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Why I Stopped Using RAG for Coding Agents (And What I Do Instead)]]></title>
            <link>https://www.clint-johnson.com/articles/coding-agents-file-exploration</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/coding-agents-file-exploration</guid>
            <pubDate>Tue, 11 Jun 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[The instinct when building a coding agent is "I need RAG to handle large codebases." The better instinct is giving the agent tools to explore code the way a senior engineer would — reading files, following imports, tracing execution.]]></description>
            <content:encoded><![CDATA[The instinct when building a coding agent is "I need RAG to handle large codebases." The better instinct is giving the agent tools to explore code the way a senior engineer would — reading files, following imports, tracing execution.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/coding-agents-file-exploration/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[React Tooling 2024: Stop Using the Wrong Shit]]></title>
            <link>https://www.clint-johnson.com/articles/react-current</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/react-current</guid>
            <pubDate>Tue, 04 Jun 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[After building 20+ React apps this year, here's what actually works and what's a waste of time. Spoiler: You're probably overengineering.]]></description>
            <content:encoded><![CDATA[After building 20+ React apps this year, here's what actually works and what's a waste of time. Spoiler: You're probably overengineering.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/react-current/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Neural Net Training Recipe That Actually Works]]></title>
            <link>https://www.clint-johnson.com/articles/neural-net-training-recipe</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/neural-net-training-recipe</guid>
            <pubDate>Wed, 17 Apr 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[I spent months chasing architecture fixes when my real problem was bad debugging hygiene. The training recipe that works — start simple, visualize everything, tune last — is the unglamorous discipline that separates working models from expensive experiments.]]></description>
            <content:encoded><![CDATA[I spent months chasing architecture fixes when my real problem was bad debugging hygiene. The training recipe that works — start simple, visualize everything, tune last — is the unglamorous discipline that separates working models from expensive experiments.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/neural-net-training-recipe/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[You Don't Need GPT-4 for That: Small Models and Edge Agents]]></title>
            <link>https://www.clint-johnson.com/articles/small-models-edge-agents</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/small-models-edge-agents</guid>
            <pubDate>Sat, 30 Mar 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[The assumption that frontier models are required for agentic function calling is wrong — and for healthcare AI, it can also be a compliance liability. Here's when a fine-tuned 7B model is the right architecture, and when it isn't.]]></description>
            <content:encoded><![CDATA[The assumption that frontier models are required for agentic function calling is wrong — and for healthcare AI, it can also be a compliance liability. Here's when a fine-tuned 7B model is the right architecture, and when it isn't.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/small-models-edge-agents/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Multi-Agent Orchestration in Practice: What I Learned Building Parallel Agent Systems]]></title>
            <link>https://www.clint-johnson.com/articles/multi-agent-orchestration</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/multi-agent-orchestration</guid>
            <pubDate>Mon, 11 Mar 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[The orchestrator/worker pattern is the key mental model for multi-agent systems. Here is how to structure orchestrators, spawn and manage workers, aggregate results, and avoid the coordination failures that will sink you.]]></description>
            <content:encoded><![CDATA[The orchestrator/worker pattern is the key mental model for multi-agent systems. Here is how to structure orchestrators, spawn and manage workers, aggregate results, and avoid the coordination failures that will sink you.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/multi-agent-orchestration/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[What It Actually Takes to Build a Real LLM Agent]]></title>
            <link>https://www.clint-johnson.com/articles/building-real-agents</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/building-real-agents</guid>
            <pubDate>Tue, 16 Jan 2024 00:00:00 GMT</pubDate>
            <description><![CDATA[Everyone's talking about agents. Few people have actually built one that works in production. Here's what the architecture papers skip: the failure modes, the memory tradeoffs, the tool design decisions that actually matter.]]></description>
            <content:encoded><![CDATA[Everyone's talking about agents. Few people have actually built one that works in production. Here's what the architecture papers skip: the failure modes, the memory tradeoffs, the tool design decisions that actually matter.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/building-real-agents/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Three Hard Truths About LLMs in Production Nobody Warned Me About]]></title>
            <link>https://www.clint-johnson.com/articles/llm-production-hard-truths</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/llm-production-hard-truths</guid>
            <pubDate>Mon, 11 Dec 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[Twelve years in ML and I still got burned. Stochasticity is a systems design problem. Your target model will be deprecated. Silent failures are worse than loud ones. Here is what production healthcare AI actually taught me.]]></description>
            <content:encoded><![CDATA[Twelve years in ML and I still got burned. Stochasticity is a systems design problem. Your target model will be deprecated. Silent failures are worse than loud ones. Here is what production healthcare AI actually taught me.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/llm-production-hard-truths/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[What AI Agents Actually Are (And What They Can't Do Yet)]]></title>
            <link>https://www.clint-johnson.com/articles/what-ai-agents-actually-are</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/what-ai-agents-actually-are</guid>
            <pubDate>Thu, 23 Nov 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[Everyone is building 'agents.' Most are just APIs with a system prompt. Here's the precise definition, what the components that actually matter are, the failure modes I've hit, and how to pick the right pattern for your problem.]]></description>
            <content:encoded><![CDATA[Everyone is building 'agents.' Most are just APIs with a system prompt. Here's the precise definition, what the components that actually matter are, the failure modes I've hit, and how to pick the right pattern for your problem.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/what-ai-agents-actually-are/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Stop Using Optional Props: Why Discriminated Unions Make Better React Components]]></title>
            <link>https://www.clint-johnson.com/articles/typescript-discriminated-unions-react</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/typescript-discriminated-unions-react</guid>
            <pubDate>Wed, 23 Aug 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[Optional props feel flexible until they ship a runtime error to production that TypeScript should have caught at compile time. Here is the pattern I now use in every component.]]></description>
            <content:encoded><![CDATA[Optional props feel flexible until they ship a runtime error to production that TypeScript should have caught at compile time. Here is the pattern I now use in every component.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/typescript-discriminated-unions-react/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[CDS Hooks: What Nobody Tells You Before You Build]]></title>
            <link>https://www.clint-johnson.com/articles/cds</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/cds</guid>
            <pubDate>Sat, 05 Aug 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[I spent months building a CDS Hooks integration that actually worked. Here's what the documentation doesn't mention.]]></description>
            <content:encoded><![CDATA[I spent months building a CDS Hooks integration that actually worked. Here's what the documentation doesn't mention.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/cds/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[The Food Truck Method: Building MVPs That Don't Suck]]></title>
            <link>https://www.clint-johnson.com/articles/tech-debt</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/tech-debt</guid>
            <pubDate>Tue, 18 Jul 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[After building dozens of MVPs, I finally figured out why most turn into dumpster fires. Here's my framework for building fast without the regret.]]></description>
            <content:encoded><![CDATA[After building dozens of MVPs, I finally figured out why most turn into dumpster fires. Here's my framework for building fast without the regret.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/tech-debt/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[React State Management & Clean SPA Architecture]]></title>
            <link>https://www.clint-johnson.com/articles/react-state</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/react-state</guid>
            <pubDate>Wed, 24 May 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[State management becomes much less important when working with SSR, but it's still important to understand the basics of state management in React.js. Chances are you may see a React.js single page application with a need for simple state management. In this article, we will explore the most popular state management tools in the React.js ecosystem: Redux, Zustand, Recoil, and the Context API, but there are many more.]]></description>
            <content:encoded><![CDATA[State management becomes much less important when working with SSR, but it's still important to understand the basics of state management in React.js. Chances are you may see a React.js single page application with a need for simple state management. In this article, we will explore the most popular state management tools in the React.js ecosystem: Redux, Zustand, Recoil, and the Context API, but there are many more.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/react-state/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[HIPAA Consent: What Engineers Actually Need to Know]]></title>
            <link>https://www.clint-johnson.com/articles/consent</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/consent</guid>
            <pubDate>Sat, 06 May 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[I spent 3 months drowning in HIPAA documentation. Here's what actually matters for engineers building healthcare apps.]]></description>
            <content:encoded><![CDATA[I spent 3 months drowning in HIPAA documentation. Here's what actually matters for engineers building healthcare apps.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/consent/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Ingesting HL7v2 with ❄️ Snowflake Part 1]]></title>
            <link>https://www.clint-johnson.com/articles/snowflake-retro</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/snowflake-retro</guid>
            <pubDate>Mon, 17 Apr 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[Snowflake is a cloud-based data warehouse system that enables organizations to securely store, query, and analyze their vast amounts of data. It has powerful capabilities for managing complex workloads and quickly adapting to changing business needs. This article will explore how Snowflake can be used as a data warehouse for healthcare organizations.]]></description>
            <content:encoded><![CDATA[Snowflake is a cloud-based data warehouse system that enables organizations to securely store, query, and analyze their vast amounts of data. It has powerful capabilities for managing complex workloads and quickly adapting to changing business needs. This article will explore how Snowflake can be used as a data warehouse for healthcare organizations.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/snowflake-retro/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[HL7v2 Healthcare Integrations with 🫏 Mulesoft]]></title>
            <link>https://www.clint-johnson.com/articles/mulesoft</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/mulesoft</guid>
            <pubDate>Thu, 30 Mar 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[In this article, we will explore how to utilize the MuleSoft Accelerator for achieving a comprehensive patient view through FHIR and HL7 v2-based integrations with EHRs and other clinical applications.]]></description>
            <content:encoded><![CDATA[In this article, we will explore how to utilize the MuleSoft Accelerator for achieving a comprehensive patient view through FHIR and HL7 v2-based integrations with EHRs and other clinical applications.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/mulesoft/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[How to ingest HL7v2 Messages into Google Cloud]]></title>
            <link>https://www.clint-johnson.com/articles/google-health</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/google-health</guid>
            <pubDate>Sun, 12 Mar 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[In this article, we will explore how to utilize the Google Cloud Healthcare API for capturing HL7v2 over MLLP, converting it to FHIR for cloud use.]]></description>
            <content:encoded><![CDATA[In this article, we will explore how to utilize the Google Cloud Healthcare API for capturing HL7v2 over MLLP, converting it to FHIR for cloud use.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/google-health/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Mirth Connect: Part 2 - Adding 🐰RabbitMQ]]></title>
            <link>https://www.clint-johnson.com/articles/rabbit-and-mirth</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/rabbit-and-mirth</guid>
            <pubDate>Wed, 22 Feb 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[Using RabbitMQ with Mirth Connect can offer several advantages, especially in complex healthcare integration scenarios. Integrations can become more durable, scalable, and easier to maintain.]]></description>
            <content:encoded><![CDATA[Using RabbitMQ with Mirth Connect can offer several advantages, especially in complex healthcare integration scenarios. Integrations can become more durable, scalable, and easier to maintain.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/rabbit-and-mirth/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Mirth Connect: Part 1 - Getting Started with 🐳 Docker]]></title>
            <link>https://www.clint-johnson.com/articles/mirth-up-and-running</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/mirth-up-and-running</guid>
            <pubDate>Fri, 03 Feb 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[Getting up and running with the industry's oldest and most widely used healthcare interoperability standard. Let's setup the popular open source integration engine Mirth Connect to send and receive HL7v2 messages.]]></description>
            <content:encoded><![CDATA[Getting up and running with the industry's oldest and most widely used healthcare interoperability standard. Let's setup the popular open source integration engine Mirth Connect to send and receive HL7v2 messages.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/mirth-up-and-running/hero.png" length="0" type="image/png"/>
        </item>
        <item>
            <title><![CDATA[Healthcare Integration For Beginners: HL7v2]]></title>
            <link>https://www.clint-johnson.com/articles/hl7v2-intro</link>
            <guid isPermaLink="false">https://www.clint-johnson.com/articles/hl7v2-intro</guid>
            <pubDate>Mon, 16 Jan 2023 00:00:00 GMT</pubDate>
            <description><![CDATA[HL7v2, is a messaging standard widely used in the healthcare industry to support clinical and administrative data communication between different software applications.]]></description>
            <content:encoded><![CDATA[HL7v2, is a messaging standard widely used in the healthcare industry to support clinical and administrative data communication between different software applications.]]></content:encoded>
            <author>clint@1putthealth.com (Clint Johnson)</author>
            <enclosure url="https://images.clint-johnson.com/articles/hl7v2-intro/hero.png" length="0" type="image/png"/>
        </item>
    </channel>
</rss>