<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Robert Terakedis</title>
    <link>https://blog.terakedis.dev/</link>
    <description>Recent content on Robert Terakedis</description>
    <generator>Hugo</generator>
    <language>en</language>
    <copyright>Copyright © Robert Terakedis; all rights reserved.</copyright>
    <lastBuildDate>Tue, 04 Aug 2026 13:02:11 -0400</lastBuildDate>
    <atom:link href="https://blog.terakedis.dev/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>leanwheel-skills, Round Three: Skip the Docs, Keep the Discipline</title>
      <link>https://blog.terakedis.dev/post/leanwheel-skills-round-three/</link>
      <pubDate>Tue, 04 Aug 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/leanwheel-skills-round-three/</guid>
      <description>&lt;p&gt;This is the third post about &lt;a href=&#34;https://github.com/rterakedis/leanwheel-skills&#34;&gt;leanwheel-skills&lt;/a&gt;. The &lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-claude-pro-workflow/&#34;&gt;first one&lt;/a&gt; covered why I built it: a token-lean port of the BMAD planning flywheel that better aligns to Claude Pro&#39;s context budget. The &lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;second one&lt;/a&gt; covered the month after: plugin packaging, a pre-planning layer, autonomous epics.&lt;/p&gt;&#xA;&lt;p&gt;Since then I&#39;ve kept running it daily on real projects. Here&#39;s what changed - including two ideas I wrote down and deliberately did &lt;em&gt;not&lt;/em&gt; build.&lt;/p&gt;&#xA;&lt;h2 id=&#34;not-everything-is-a-project&#34;&gt;Not everything is a project&lt;/h2&gt;&#xA;&lt;p&gt;A recent chat with a coworker led me to a realization: leanwheel assumed every piece of work lived inside a project. Every lane started from &lt;code&gt;docs/prd.md&lt;/code&gt; and &lt;code&gt;docs/architecture.md&lt;/code&gt;, every story traced back to an epic, every change updated the docs. That&#39;s part of what led me to BMAD to begin with.  As a prolific documentarian, I &lt;em&gt;loved&lt;/em&gt; all the documentation built and consumed by BMAD.  That&#39;s the right shape for building a large project from nothing. It&#39;s the wrong shape for most of what a working week actually looks like: a fix in a coworker&#39;s repo, a one-off script in a folder that will never have a PRD, a quick change in a fresh clone you&#39;ll touch exactly once.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Knowledge Pitfalls in Productivity - Lessons Learned</title>
      <link>https://blog.terakedis.dev/post/pitfalls-to-productivity---lessons-learned/</link>
      <pubDate>Wed, 29 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/pitfalls-to-productivity---lessons-learned/</guid>
      <description>&lt;p&gt;You may have heard the saying &amp;quot;Thinking out loud?&amp;quot;   I&#39;m the type of person that thinks on paper:  I flowchart, I list, I document!  I&#39;m also many years into my career and find it hard to believe that organizations &lt;em&gt;still&lt;/em&gt; struggle with the same Knowledge Management problems as I saw early in my career. I&#39;m sure you&#39;ve seen many of these in your own job:&lt;/p&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;Knowledge documented, but noone can find it easily (Knowledge App Sprawl).&lt;/li&gt;&#xA;&lt;li&gt;Knowledge disappearing (due to attrition).&lt;/li&gt;&#xA;&lt;li&gt;Knowledge maintenance issues&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&lt;p&gt;It&#39;s a real problem.  I saw an infographic on &lt;a href=&#34;https://www.getguru.com/knowledge-driven-culture&#34;&gt;Guru&#39;s site&lt;/a&gt; saying that 82% of leaders use five or more sources of knowledge throughout the day. Yes, I know that&#39;s a vendor site trying to sell a product, but that&#39;s still bonkers... Frankly, I don&#39;t know how businesses function when knowledge is so fragmented throughout the organization.  I&#39;m by no means an expert in Knowledge Management, but it&#39;s become a side study of mine over the past few years.  Maybe it&#39;s a lofty goal, but I have to think knowledge disorganization is fixable.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Before You Ship an App: The Boring Stuff Nobody Warns You About</title>
      <link>https://blog.terakedis.dev/post/before-you-ship-an-app/</link>
      <pubDate>Tue, 14 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/before-you-ship-an-app/</guid>
      <description>&lt;p&gt;I went into building an app with a very simple mental model: write the code, hit publish, done. The App Store was the finish line. Everything hard was going to be in the code.&lt;/p&gt;&#xA;&lt;p&gt;That turned out to be backwards. The code was the part I understood. What I didn&#39;t understand was that before a single customer can hand you a dollar, you&#39;re standing up a small business, taking on real liability, and making privacy decisions that are a lot harder to unwind later than they are to get right up front. Privacy, &lt;a href=&#34;https://blog.terakedis.dev/post/privacy-certifications-systems-design/&#34;&gt;as I&#39;ve come to see it&lt;/a&gt;, is an architecture problem, not a checkbox you tick at the end, and that&#39;s exactly why the decisions you make now are the ones you&#39;ll be living with.&lt;/p&gt;</description>
    </item>
    <item>
      <title>leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math</title>
      <link>https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/</link>
      <pubDate>Mon, 13 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/</guid>
      <description>&lt;p&gt;A few weeks ago I wrote about &lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-claude-pro-workflow/&#34;&gt;building leanwheel-skills&lt;/a&gt;, a stripped-down port of the BMAD planning flywheel that fits inside Claude Pro&#39;s context budget. The pitch was simple: keep the structured PRD → architecture → epics → stories → dev → review loop, cut the overhead that doesn&#39;t earn its token cost, come out around 55% lighter.&lt;/p&gt;&#xA;&lt;p&gt;That was the starting point. I&#39;ve been using it on real projects since, and it grew. A lot. The interesting part is that it got noticeably more capable while staying lean - the token savings actually went &lt;em&gt;up&lt;/em&gt;, not down, even as I added whole new phases. Here&#39;s what changed.&lt;/p&gt;</description>
    </item>
    <item>
      <title>From Docs-as-Code to Docs-in-Code</title>
      <link>https://blog.terakedis.dev/post/from-docs-as-code-to-docs-in-code/</link>
      <pubDate>Fri, 26 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/from-docs-as-code-to-docs-in-code/</guid>
      <description>&lt;p&gt;If you&#39;ve worked in a modern engineering org, you already know docs-as-code. You write documentation in markdown, you keep it in version control, it renders nicely in GitHub or GitLab, and it gets reviewed in the same pull request as the change it describes. The win is that docs live next to the code and move through the same gates. No separate publishing step, no wiki that drifts out of date the moment someone merges.&lt;/p&gt;</description>
    </item>
    <item>
      <title>How 46 Hours Inside macOS Logs Made Me a Better Architect</title>
      <link>https://blog.terakedis.dev/post/debugging-as-architectural-discipline/</link>
      <pubDate>Thu, 25 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/debugging-as-architectural-discipline/</guid>
      <description>&lt;p&gt;Back in 2020, I finally sat down and pulled every note I had on troubleshooting macOS into a single guide. It took about 46 hours of fingers on keys - typing, formatting, and testing every command before it went out the door. It shipped on VMware&#39;s TechZone and got more kudos than I expected for something that had sat on my to-do list way too long.&lt;/p&gt;&#xA;&lt;p&gt;Platforms change hands, and through a couple of acquisitions the original byline didn&#39;t follow the content around. That&#39;s how it goes. The expertise is still mine, though, and that&#39;s the part worth preserving - so here it is. I wrote up the original experience at the time in &lt;a href=&#34;https://blog.terakedis.dev/post/troubleshooting-macos-management-with-workspace-one/&#34;&gt;Troubleshooting macOS Management with Workspace ONE&lt;/a&gt;. This post is about what outlasted the platform.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Building a Leaner AI Development Workflow for Claude Pro</title>
      <link>https://blog.terakedis.dev/post/leanwheel-skills-claude-pro-workflow/</link>
      <pubDate>Mon, 15 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/leanwheel-skills-claude-pro-workflow/</guid>
      <description>&lt;p&gt;Earlier this year, I started using &lt;a href=&#34;https://github.com/bmad-method/bmad&#34;&gt;BMAD&lt;/a&gt; as a structured way to build software with Windsurf at work. The basic idea is sound: write planning docs first (PRD, architecture, epic breakdown), then give the AI one well-scoped story at a time to implement. Instead of one giant &amp;quot;build me this app&amp;quot; prompt that wanders off course, you get repeatable cycles with clear checkpoints.&lt;/p&gt;&#xA;&lt;p&gt;It worked well enough that I wanted to use it on everything, including some hobby projects at home. Then I ran into the context budget like a brick wall.&lt;/p&gt;</description>
    </item>
    <item>
      <title>What Privacy Certifications Taught Me About Systems Design</title>
      <link>https://blog.terakedis.dev/post/privacy-certifications-systems-design/</link>
      <pubDate>Mon, 15 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/privacy-certifications-systems-design/</guid>
      <description>&lt;p&gt;I got my CIPP/E and CIPM because my employer required it. That&#39;s the honest version. I wasn&#39;t chasing credentials or pivoting to a privacy career - it was on the list of things I needed to get done.&lt;/p&gt;&#xA;&lt;p&gt;What I didn&#39;t expect was how much it changed the way I think about architecture.&lt;/p&gt;&#xA;&lt;p&gt;Before the certifications, I thought about privacy the way most engineers do: a checklist item somewhere near the end of a project. &amp;quot;Do we have a privacy policy? Do we have consent forms? Are we logging the right things?&amp;quot; Check, check, check. Done.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Designing an AI Pipeline That Works Everywhere, All the Time</title>
      <link>https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/</link>
      <pubDate>Sun, 14 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/</guid>
      <description>&lt;p&gt;I wrote about the &lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;matching logic behind this system&lt;/a&gt; in an earlier post. That post covers the five-layer architecture: intake, vector search, LLM scoring, threshold routing, and feedback. What it doesn&#39;t cover is the part that took just as much thought: making the whole thing reliable enough that the team can count on it regardless of when or where they&#39;re working.&lt;/p&gt;&#xA;&lt;p&gt;Getting a prototype working on a Friday afternoon is one problem. Getting a system to reliably respond to events around the clock, across a globally distributed team, is a different problem. That&#39;s what this post is about.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Why I Stopped Treating LLMs as Magic and Started Treating Them as Components</title>
      <link>https://blog.terakedis.dev/post/llms-are-not-magic/</link>
      <pubDate>Sat, 16 May 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/llms-are-not-magic/</guid>
      <description>&lt;p&gt;I recently got the opportunity to build out an AI-powered feature-matching system.  The idea had already been floated, and the assumption was that it&#39;d be easy - give an LLM a spreadsheet of ideas and features, let it give you back the list of matches.   Feed it the customer&#39;s idea, get back a match. Easy.   I initially made the same assumption a lot of teams make: I treated the LLM as &lt;em&gt;the solution&lt;/em&gt;.&lt;/p&gt;</description>
    </item>
    <item>
      <title>SSO Is Not a Flip Switch</title>
      <link>https://blog.terakedis.dev/post/sso-is-not-a-flip-switch/</link>
      <pubDate>Fri, 15 May 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/sso-is-not-a-flip-switch/</guid>
      <description>&lt;p&gt;When we started modernizing the LMS infrastructure, the first thing someone asked was: &amp;quot;Can we just add SSO to the LMS?&amp;quot;&lt;/p&gt;&#xA;&lt;p&gt;It&#39;s a reasonable question. It sounds like a small ask - flip a switch, wire up a login button, done. But the moment you start pulling on that thread, you find out it&#39;s not a switch at all. It&#39;s a decision about where identity lives across every system you own, and getting it wrong means you end up with the same mess you started with, just with a single sign-on button in front of it.&lt;/p&gt;</description>
    </item>
    <item>
      <title>The Six-Figure Spreadsheet: How Manual Provisioning Became a Liability</title>
      <link>https://blog.terakedis.dev/post/the-six-figure-spreadsheet/</link>
      <pubDate>Sat, 18 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/the-six-figure-spreadsheet/</guid>
      <description>&lt;p&gt;Last year I got pulled into a project where the team was drowning in spreadsheets. Not the good kind, but rather the unfortunate kind: manual user provisioning lists, copy-paste operations between systems, and a recurring cloud bill that kept climbing because of over-provisioned resources sitting idle.&lt;/p&gt;&#xA;&lt;p&gt;The kicker? Along with all the manual inefficiences, we were burning six figures a year in overprovisioned cloud spend. And that was before we even talked about the staff time or the risk of things breaking when a step in the process was missed.&lt;/p&gt;</description>
    </item>
    <item>
      <title>How to host blog images for free with Backblaze B2 and Cloudflare</title>
      <link>https://blog.terakedis.dev/post/b2-cloudflare-image-hosting/</link>
      <pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/b2-cloudflare-image-hosting/</guid>
      <description>&lt;p&gt;When I migrated this blog to Hugo, I had to figure out where images would live. My original thought was simple enough: store them in the GitHub repo. That worked fine for six months. Then I started thinking about what happens when I have hundreds of posts with images, and the repo turns into a photo dump.&lt;/p&gt;&#xA;&lt;p&gt;I looked at Cloudflare Images next. It seemed obvious - I&#39;m already paying for Cloudflare for DNS and SSL. But the pricing pushed me away: the transform API costs add up if you&#39;re generating variants (WebP, different sizes), and I&#39;d be locking myself into Cloudflare&#39;s image manipulation if I ever wanted to migrate. That didn&#39;t feel like room to grow.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Setting up our Calendar Max</title>
      <link>https://blog.terakedis.dev/post/setting-up-skylight-calendar-max/</link>
      <pubDate>Sun, 11 May 2025 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/setting-up-skylight-calendar-max/</guid>
      <description>&lt;p&gt;If you’ve ever struggled to keep your family’s schedule in sync - or just wanted a little less chaos in your mornings - the Calendar Max might be your new best friend. After seeing it pop up in a Mother’s Day promo, we decided to give it a shot. Spoiler: our living room wall makes me smile every time I walk by, and even our kids are excited to check what’s coming up next!&lt;/p&gt;</description>
    </item>
    <item>
      <title>When the Workflow Is the Bug: Rebuilding Multi-Team Processes in Jira</title>
      <link>https://blog.terakedis.dev/post/when-the-workflow-is-the-bug/</link>
      <pubDate>Mon, 10 Feb 2025 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/when-the-workflow-is-the-bug/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I was asked to help figure out how we could be more proactive in notifying customers of changes. My first thought: &amp;quot;We already have a release process - we just need to layer communication on top.&amp;quot; Turns out it was more complicated. The problem was that multiple teams (engineering, product management, and customer support) needed to coordinate on &lt;em&gt;infrastructure and operational changes&lt;/em&gt; (changes outside the normal software development cycle), but there was no consistent, repeatable process to make that happen. Worse, customers were finding out about changes before anyone inside had a plan to notify them.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Postman Pre-Request Script for Docebo API</title>
      <link>https://blog.terakedis.dev/post/postman-pre-request-script-for-docebo-api/</link>
      <pubDate>Wed, 22 Jan 2025 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/postman-pre-request-script-for-docebo-api/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I&#39;ve recently been involved with a project at work to modernize the processes around Docebo, our Learning Management System (LMS). A significant portion of this project involves workflow automation and passing data/events between multiple systems. We accomplish this automation using webhooks and a number of API calls (which you can view at &lt;a href=&#34;https://doceboapi.docebosaas.com/api-browser/&#34;&gt;Docebo&#39;s API Browser&lt;/a&gt;).  Early in the process, I stumbled onto the Docebo Community post about &lt;a href=&#34;https://community.docebo.com/integrations-apis-45/after-the-community-coaching-session-using-postman-9897&#34;&gt;Using Postman&lt;/a&gt; (and the on-demand &lt;a href=&#34;https://community.docebo.com/events/on-demand-community-coaching-session-using-postman-with-the-api-97&#34;&gt;Community Coaching Session&lt;/a&gt;) which contained a downloadable Postman API collection.  This was a GREAT starting point, but the collection used an interactive login to get oAuth2 tokens. I wanted to automate the authentication process and avoid the interactive logins when I was testing against our sandbox LMS instance.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Getting Started: Learning Systems Thinking</title>
      <link>https://blog.terakedis.dev/gettingstarted/</link>
      <pubDate>Sat, 21 Sep 2024 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/gettingstarted/</guid>
      <description>&lt;p&gt;People ask me how to get better at their work - whether it&#39;s device management, automation, architecture, or just understanding why complex systems behave the way they do. The answer isn&#39;t usually &amp;quot;read the docs&amp;quot; (though that helps). It&#39;s about developing systems thinking. Here are resources that changed how I approach problems.&lt;/p&gt;&#xA;&lt;h2 id=&#34;systems-thinking-start-with-the-goal&#34;&gt;Systems Thinking: Start with &amp;quot;The Goal&amp;quot;&lt;/h2&gt;&#xA;&lt;p&gt;If you read one business book, read &lt;a href=&#34;https://www.amazon.com/Goal-Process-Ongoing-Improvement/dp/0884271951&#34;&gt;&lt;strong&gt;The Goal&lt;/strong&gt; by Eliyahu Goldratt&lt;/a&gt;. It&#39;s a novel about manufacturing, but it&#39;s really about how to think about constraints, bottlenecks, and outcomes.&lt;/p&gt;</description>
    </item>
    <item>
      <title>About me</title>
      <link>https://blog.terakedis.dev/about/</link>
      <pubDate>Tue, 19 Mar 2024 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/about/</guid>
      <description>&lt;p&gt;I&#39;m Robert Terakedis, a Principal Solutions Architect based in North Carolina. My background spans Azure cloud architecture, AI/LLM systems, enterprise automation, and about a decade of Apple platform work before that. I started in datacenter infrastructure at a bank, which taught me to think in terms of scale and operational reliability before I&#39;d ever heard the term &amp;quot;cloud native.&amp;quot;&lt;/p&gt;&#xA;&lt;p&gt;Most recently at OneTrust, I&#39;ve been building AI-powered automation systems in Python on Azure - things like automated idea/feature matching pipelines with vector embeddings, enterprise integrations between platforms that were never designed to talk to each other, and product analytics instrumentation. Before that, I was the Apple platform SME at VMware, where I built reference architectures and ran certification programs for enterprise engineers. That role is where I learned to write for a technical practitioner audience, which I&#39;ve been doing ever since.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Portfolio</title>
      <link>https://blog.terakedis.dev/portfolio/</link>
      <pubDate>Tue, 19 Mar 2024 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/portfolio/</guid>
      <description>&lt;h2 id=&#34;whats-here&#34;&gt;What’s here&lt;/h2&gt;&#xA;&lt;p&gt;This page pulls together my writing on the topics I care most about right now: &lt;strong&gt;Azure cloud architecture&lt;/strong&gt;, &lt;strong&gt;AI/LLM systems&lt;/strong&gt;, &lt;strong&gt;Python automation&lt;/strong&gt;, and &lt;strong&gt;systems thinking&lt;/strong&gt;.&lt;/p&gt;&#xA;&lt;p&gt;I write through blog posts rather than static samples because posts are more honest - they show how I actually approached a problem, including the parts that didn’t work the first time. If you want to get a sense of how I think, the posts below are a better signal than a resume bullet.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Recommendations</title>
      <link>https://blog.terakedis.dev/recommendations/</link>
      <pubDate>Tue, 19 Mar 2024 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/recommendations/</guid>
      <description>&lt;p&gt;These come from people I&#39;ve worked alongside across architecture, process design, and technical enablement roles. They&#39;re a sample. For the full list of 35+ recommendations, see &lt;a href=&#34;https://www.linkedin.com/in/terakedis/&#34;&gt;my LinkedIn profile&lt;/a&gt;.&lt;/p&gt;&#xA;&lt;div class=&#34;rec-grid&#34;&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote rec-quote--clamp&#34;&gt;&#xA;        I had the pleasure of working alongside Rob, and he&#39;s truly one of the most talented professionals I&#39;ve had the privilege of collaborating with. His ability to take complex technical requirements and seamlessly translate them into simplified, actionable functional requirements is nothing short of remarkable. Rob is the epitome of detail-oriented and organized. Every project he touches benefits from his meticulous attention to detail and superb documentation skills, ensuring clarity and alignment across all stakeholders. What truly sets Rob apart, though, is his infectious positive energy and sense of humor. He brings an uplifting dynamic to the team, creating an environment where collaboration thrives. No matter how challenging the situation, he knows how to keep morale high and focus sharp. Anyone who has the chance to work with Rob will quickly realize how invaluable he is!&#xA;      &lt;/blockquote&gt;&lt;details class=&#34;rec-more&#34;&gt;&#xA;        &lt;summary&gt;&lt;span class=&#34;rec-more__more&#34;&gt;Read full recommendation&lt;/span&gt;&lt;span class=&#34;rec-more__less&#34;&gt;Show less&lt;/span&gt;&lt;/summary&gt;&#xA;      &lt;/details&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/quy-trinh/&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;Quy Trinh&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Senior Consultant · OneTrust&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote rec-quote--clamp&#34;&gt;&#xA;        I worked with Robert first as a peer and later as his direct manager. Robert has been an absolute pleasure to work with. From day one he was engaged and driven to learn any technology he could get his hands on. When we first began working together we were a small team and stretched thin and Rob stepped up to take on responsibilities outside his area of expertise with such determination that he would eventually go on to take full responsibility as a subject matter expert. He is the kind of individual whose absence is immediately felt when he is not in the office. Rob is always thinking of others and strives to improve himself and those around him. He listens to customer and peer issues intently, does his research and returns with well thought out and detailed responses, anticipating most questions that may arise and answering those at the same time. I have complete trust when assigning Robert a task and have thoroughly enjoyed working with him. Without hesitation I would recommend him in any technology or people related role.&#xA;      &lt;/blockquote&gt;&lt;details class=&#34;rec-more&#34;&gt;&#xA;        &lt;summary&gt;&lt;span class=&#34;rec-more__more&#34;&gt;Read full recommendation&lt;/span&gt;&lt;span class=&#34;rec-more__less&#34;&gt;Show less&lt;/span&gt;&lt;/summary&gt;&#xA;      &lt;/details&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/jeffeharris&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;Jeff Harris&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Manager · VMware&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote rec-quote--clamp&#34;&gt;&#xA;        I had the privilege of seeing Rob work a lot of his magic in a project we worked through from September 2023 until March 2024. He&#39;s got a humble demeanor but is a stacked and MVP-level contributor! When coming up with the workflow for a new process, Rob proactively reached out to stakeholders to understand the needs and use cases, created a process flow diagram for everyone to align on, and worked quickly with our Jira Admin to stand up the Jira Service Management form. From start to finish he functioned like a real team player - he knew when to delegate, clearly defined who owned what at each step, and questioned band-aid solutions in favor of scalable ones. That kept us on track and cohesive across a team spanning Product, Engineering, CS, PS, Support, and Partners.&#xA;      &lt;/blockquote&gt;&lt;details class=&#34;rec-more&#34;&gt;&#xA;        &lt;summary&gt;&lt;span class=&#34;rec-more__more&#34;&gt;Read full recommendation&lt;/span&gt;&lt;span class=&#34;rec-more__less&#34;&gt;Show less&lt;/span&gt;&lt;/summary&gt;&#xA;      &lt;/details&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/rachskim&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;Rachel Kim&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Program Manager · OneTrust&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote&#34;&gt;&#xA;        I had the pleasure of working with Rob on a high priority project at OneTrust. Rob&#39;s technical knowledge, ability to document processes and thoroughness is extremely commendable. Rob has a fantastic ability (and patience) to explain technical aspects of systems in a way that someone who is &#39;not so technical&#39; would understand. I&#39;ll forever be grateful for the knowledge I obtained and learned whilst working alongside Rob! He&#39;s a rockstar in his field!&#xA;      &lt;/blockquote&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/keeleylucraft&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;Keeley Lucraft&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Senior Consultant · OneTrust&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote&#34;&gt;&#xA;        I had the pleasure of working with Rob during my time at VMware. Not only is Rob extremely knowledgeable about the VMware ecosystem and the role of Mac within it, he&#39;s also always eager to help out, answer questions, and otherwise share that knowledge. Rob works well within diverse teams across organizational segments and verticals to deliver an optimal experience for customers.&#xA;      &lt;/blockquote&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/smrissland&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;Sonja Rissland&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Product Manager · VMware&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote&#34;&gt;&#xA;        Rob&#39;s ability to fully understand a complex platform, break it down into manageable parts and then deliver enablement for each piece is unrivalled. I consider myself incredibly fortunate to be able to call upon Rob for advice and direction, and the quality of his work makes me strive to continuously improve. The passion Rob puts into all of his work benefits the community as a whole and he is a fantastic colleague to work with.&#xA;      &lt;/blockquote&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/adampmatthews&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;Adam Matthews&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Sales Engineer · VMware&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote&#34;&gt;&#xA;        Rob is a real Apple (iOS, macOS, and tvOS) guru. In his role in Technical Marketing he has proved to be invaluable in enabling the WW EUC Solution Engineers to advise customers on the benefits of using Workspace ONE for configuring and managing their Apple devices. He produces high-quality presentations, blogs, and podcasts as well as always being keen to answer any questions we may have. He truly is an amazing team player.&#xA;      &lt;/blockquote&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/james-murray-970b515&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;James Murray&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Sales Engineer · VMware&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&lt;figure class=&#34;rec-card&#34;&gt;&#xA;      &lt;blockquote class=&#34;rec-quote&#34;&gt;&#xA;        Having worked with Rob for just under two years, I couldn&#39;t be more grateful to have met someone. Rob&#39;s kind and friendly nature make him easy to approach, coupled with his knowledge and quick responses to questions he is an invaluable member of our team. The material that Rob produces is widely used not just within our own organisation but across many, making him a trusted member across multiple communities.&#xA;      &lt;/blockquote&gt;&#xA;      &lt;figcaption class=&#34;rec-cite&#34;&gt;&#xA;        &lt;a class=&#34;rec-name&#34; href=&#34;https://www.linkedin.com/in/joseph-ms-mcdonald&#34; rel=&#34;noopener&#34; target=&#34;_blank&#34;&gt;Joe McDonald&lt;/a&gt;&#xA;        &lt;span class=&#34;rec-meta&#34;&gt;Sales Engineer · VMware&lt;/span&gt;&#xA;      &lt;/figcaption&gt;&#xA;    &lt;/figure&gt;&#xA;&lt;/div&gt;</description>
    </item>
    <item>
      <title>HomePod Mini Ignoring Family Explicit Content Settings</title>
      <link>https://blog.terakedis.dev/post/homepod-mini-ignoring-family-explicit-content-settings/</link>
      <pubDate>Wed, 04 Jan 2023 22:28:08 -0500</pubDate>
      <guid>https://blog.terakedis.dev/post/homepod-mini-ignoring-family-explicit-content-settings/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Sudden panic set in as I walked by my child&#39;s room and heard the lyric from &lt;em&gt;Monsters&lt;/em&gt; by &lt;em&gt;All Time Low&lt;/em&gt; -- &amp;quot;I don&#39;t mind while you f**k up my life.&amp;quot;  Yes, I just had the realization that the critical parental controls on the Homepod Mini&#39;s in my children&#39;s rooms were no longer working as intended.  This set off a multi-hour troubleshooting session, with some involvement from AppleCare, that got things working until it didn&#39;t (again).  As it usually goes, &amp;quot;It Just Works&amp;quot;... until it doesn&#39;t.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Apple&#39;s Missing Hardware - A KVM Switch</title>
      <link>https://blog.terakedis.dev/post/apples-missing-hardware-kvm-switch/</link>
      <pubDate>Fri, 25 Nov 2022 15:40:35 -0500</pubDate>
      <guid>https://blog.terakedis.dev/post/apples-missing-hardware-kvm-switch/</guid>
      <description>&lt;p&gt;Let me start by saying that I&#39;m assuming many of you out there are in a situation quite similar to my own.  This situation; having a personal Mac &lt;em&gt;and&lt;/em&gt; a work Mac (and maybe an iPad Pro thrown into that mix also). But if you&#39;re also like me, you have limited space on your desk and don&#39;t want to be surrounded by monitors!  This leads me to my most potent complaint about Apple hardware; the lack of a KVM switch.&lt;/p&gt;</description>
    </item>
    <item>
      <title>My First Dive into BrickLink Studio and a Brick Sketch Template</title>
      <link>https://blog.terakedis.dev/post/bricklink-studio-afol-brick-sketch-template/</link>
      <pubDate>Mon, 27 Jun 2022 15:20:19 -0400</pubDate>
      <guid>https://blog.terakedis.dev/post/bricklink-studio-afol-brick-sketch-template/</guid>
      <description>&lt;p&gt;I was a total &amp;quot;builder&amp;quot; when I was young.  I loved pretty much anything that let me create my own structures:  Construx, Lego, Lincoln Logs, Marble Runs, etc.   Looking back, it really ended up being Lego that I most enjoyed playing with.   Back in my Lego heyday, Magtron and Blacktron were the sets most captivating to me.  I was fascinated by space (thanks grandparents for the trip to Kennedy Space Center).  Some of my favorite sets (which I still have some of the pieces):&lt;/p&gt;</description>
    </item>
    <item>
      <title>Ensuring Domain Name Survival</title>
      <link>https://blog.terakedis.dev/post/ensuring-domain-name-survival/</link>
      <pubDate>Fri, 24 Jun 2022 21:54:25 -0400</pubDate>
      <guid>https://blog.terakedis.dev/post/ensuring-domain-name-survival/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I was listening to the &lt;a href=&#34;https://twit.tv/posts/transcripts/security-now-episode-885-transcript&#34;&gt;Security Now Podcast&lt;/a&gt; the other day, and a GREAT topic came up that I wanted to write about.  That topic, as covered by Leo and Steve, involves ownership (specifically registrants) of Domain Name System (DNS) records.  I know from past experience, domain &amp;quot;ownership&amp;quot; never seems to be front-of-mind for many folks deciding to start a website.  Whether you&#39;re simply looking to blog, or starting a new business, the focus tends to be on the outcome (&amp;quot;I have a new website&amp;quot;) than the path to get there.   In this article, I want to cover some of the common ways domains are registered and how you can protect your domain name should something ever happen.&lt;/p&gt;</description>
    </item>
    <item>
      <title>New Job in the Privacy Industry at OneTrust</title>
      <link>https://blog.terakedis.dev/post/new-job-industry-privacy-onetrust/</link>
      <pubDate>Fri, 24 Jun 2022 21:54:25 -0400</pubDate>
      <guid>https://blog.terakedis.dev/post/new-job-industry-privacy-onetrust/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Lots of changes afoot in my professional life. This past February, I joined the ranks of many folks in the &lt;em&gt;Great Resignation&lt;/em&gt; and made a career shift. With my new role, I&#39;m working for OneTrust -- a company that creates software to help other companies manage their privacy and regulatory compliance.  It&#39;s been a total shock to the system --&amp;gt; completely new vocabulary, new verticals, everything. To say I&#39;m feeling impostor syndrome would be an understatement, but I welcome the opportunity to get out of my comfort zone and learn something new.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Change From Beautifulhugo to Clarity Theme</title>
      <link>https://blog.terakedis.dev/post/change-from-beautifulhugo-to-clarity-theme/</link>
      <pubDate>Fri, 11 Feb 2022 12:24:02 -0500</pubDate>
      <guid>https://blog.terakedis.dev/post/change-from-beautifulhugo-to-clarity-theme/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Since I first started this blog, I was using a theme called &lt;a href=&#34;https://github.com/halogenica/beautifulhugo&#34;&gt;BeautifulHugo&lt;/a&gt;.  The theme was great and for the most part visually appealing.  However, there were some changes I was hoping to see made to the theme that never really manifested.   I had even contributed back some of my own changes, but it appeared the theme had been abandoned by it&#39;s creator.  I set about looking for a new theme and stumbled onto &lt;a href=&#34;https://github.com/chipzoller/hugo-clarity&#34;&gt;hugo-clarity&lt;/a&gt;.  Not only was this theme visually appealing to me, but I could see regular changes being made to the theme.  Yes, a theme still under active maintenance!&lt;/p&gt;</description>
    </item>
    <item>
      <title>Mapping Concepts from Munki to WS1 UEM</title>
      <link>https://blog.terakedis.dev/post/mapping-concepts-from-munki-to-workspace-one-uem/</link>
      <pubDate>Mon, 23 Aug 2021 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/mapping-concepts-from-munki-to-workspace-one-uem/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Since Workspace ONE UEM version 9.3, VMware has included the open-source &lt;a href=&#34;https://github.com/munki/munki&#34;&gt;munki&lt;/a&gt; binaries in the Workspace ONE Intelligent Hub for macOS.  The intent of this integration was to give macOS admins the ability to distribute 3rd-party non-store macOS apps, without the need to host any backing instrastructure for munki.  Because this integration was meant to give admins some of the commonly used functionality, it was not integrated in such a way to include support for ALL munki&#39;s functionality.   This has left some confusion in the community, particularly with Workspace ONE UEM administrators with extensive munki knowledge/background.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Best Practices for Apple Admins in Workspace ONE UEM</title>
      <link>https://blog.terakedis.dev/post/best-practices-for-apple-admins-in-workspace-one-uem/</link>
      <pubDate>Thu, 15 Jul 2021 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/best-practices-for-apple-admins-in-workspace-one-uem/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I was recently reading a post about &lt;a href=&#34;https://macostechs.com/2021/05/what-to-do-when-you-have-to-lay-off-your-jamf-administrator/&#34;&gt;What to do when you have to lay off your Jamf administrator&lt;/a&gt;, and it got me thinking.  The &lt;a href=&#34;https://docs.vmware.com/en/VMware-Workspace-ONE-UEM/2102/UEM_Recommended_Architecture/GUID-AWT-RECARCH-INTRO.html&#34;&gt;Workspace ONE UEM documentation&lt;/a&gt; generally specifies what you &lt;strong&gt;need&lt;/strong&gt; from a software and hardware perspective in pre-requisites.  That said, over the years I&#39;ve come to know a few unwritten (or written but obscure) best practices for setting up Workspace ONE UEM to manage Apple devices.  Hopefully you find this post helpful, but I welcome any comments and feedback!&lt;/p&gt;</description>
    </item>
    <item>
      <title>Great *Free* Resources to Learn GitHub</title>
      <link>https://blog.terakedis.dev/post/great-free-resources-to-learn-github/</link>
      <pubDate>Fri, 19 Mar 2021 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/great-free-resources-to-learn-github/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Source control used to be a scary, odd thing for me back in college and early in my career.   Branching, merging, commits... all the buzzwords I remember from my early days dealing with Microsoft Team Foundation Server (TFS) and Subversion.  Admittedly, I never felt like I spent the time required to understand how it all worked.  Fast forward to mid-2017, when a group of us within VMware needed to share some of our tips and tricks with the broader VMware community.  The &lt;a href=&#34;https://github.com/vmware-samples/euc-samples&#34;&gt;EUC-Samples&lt;/a&gt; repository was born on GitHub, and my journey like Alice down the rabbit hole began.  Over the years, as we&#39;ve encouraged participation and knowledge sharing amongst those within the community, I still see a hesitation to learn git/GitHub and contribute to the repository.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Converting Jamf Custom Schema JSON for Workspace ONE UEM</title>
      <link>https://blog.terakedis.dev/post/converting-jamf-custom-schema-json-workspace-one-uem/</link>
      <pubDate>Tue, 09 Mar 2021 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/converting-jamf-custom-schema-json-workspace-one-uem/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Periodically, I see app vendors providing custom JSON schema files to help build app-specific configuration profiles for MDM (specifically Jamf).  Workspace ONE UEM supports app-specific configuration, but currently via &lt;a href=&#34;https://github.com/vmware-samples/euc-samples/blob/master/macOS-Samples/CustomXMLProfiles/CustomXMLTemplate.md&#34;&gt;Custom Settings in an XML format&lt;/a&gt;.  While many vendors also suppply a custom mobileconfig file or Custom Settings dictionary that can be used with Workspace ONE UEM, I hope in this post to show how any Workspace ONE admin can manually convert a Custom Schema JSON file to Custom Settings XML.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Troubleshooting macOS Management with Workspace ONE</title>
      <link>https://blog.terakedis.dev/post/troubleshooting-macos-management-with-workspace-one/</link>
      <pubDate>Tue, 27 Oct 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/troubleshooting-macos-management-with-workspace-one/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Short post today - just to cover some thoughts on my most recent asset published to &lt;a href=&#34;https://techzone.vmware.com&#34;&gt;TechZone&lt;/a&gt;.   I finally sat down and dug out all my notes on troubleshooting macOS and put them all together into a single, comprehensive &lt;a href=&#34;https://techzone.omnissa.com/troubleshooting-macos-management-workspace-one-operational-tutorial#overview&#34;&gt;macOS Troubleshooting Guide&lt;/a&gt;.  This thing was the result of almost 46 hours of fingers on keys: typing, formatting, and testing.  I truly hope you all get some value out of it, and do feel free to send me feedback if you&#39;d like it extended and/or notice something missing.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Updated GitHub Actions to Publish Hugo Site From Private to Public Repo</title>
      <link>https://blog.terakedis.dev/post/updated-github-actions-publish-private-hugo-repo-to-public-pages-site/</link>
      <pubDate>Tue, 27 Oct 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/updated-github-actions-publish-private-hugo-repo-to-public-pages-site/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;When I restarted my blogging journey last year, I went with Hugo to generate a static website hosted as a GitHub Pages site.  As mentioned, Blogger and WordPress always suffered recurring problems, and maintenance with WordPress still turned into a time suck due to its complexity.  By comparison, GitHub has been a nearly painless hosting provider, and the way I&#39;ve configured it has allowed me to keep drafts hidden by staging in a private repository.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Deploying NoMAD over VMware Per-App Tunnel</title>
      <link>https://blog.terakedis.dev/post/deploying-nomad-over-vmware-per-app-tunnel/</link>
      <pubDate>Fri, 16 Oct 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/deploying-nomad-over-vmware-per-app-tunnel/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Some folks recently reached out to me asking for help figuring out how to route &lt;a href=&#34;https://nomad.menu/support/&#34;&gt;NoMAD&lt;/a&gt; traffic over VMware Tunnel.   Basically, the ask was to use Per-App Tunneling to give NoMAD the ability to obtain Kerberos Tickets and Sync AD passwords without being directly on the Enterprise network.  If you&#39;re familiar with my previous article about the &lt;a href=&#34;https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn-followup/&#34;&gt;macOS Catalina Kerberos SSO Plugin&lt;/a&gt;, you&#39;ll know that Apple&#39;s built-in functionality in Catalina doesn&#39;t work well with Per-App Tunneling.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Modify ZShell Defaults in macOS</title>
      <link>https://blog.terakedis.dev/post/modify-zshell-defaults/</link>
      <pubDate>Mon, 21 Sep 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/modify-zshell-defaults/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I&#39;m finding myself using the Terminal quite a bit more in my job. I spent a few minutes over the past few days looking at different ways to make the default terminal layout in macOS better.  While there are many plugins out there for doing this (&lt;a href=&#34;https://github.com/ohmyzsh/ohmyzsh&#34;&gt;Oh-My-Zsh&lt;/a&gt;), I wanted to do something a little more straightforward.&lt;/p&gt;</description>
    </item>
    <item>
      <title>macOS Big Sur and Kerberos SSO via Per-App Tunnel</title>
      <link>https://blog.terakedis.dev/post/macos-bigsur-kerberos-sso-over-vpn/</link>
      <pubDate>Fri, 14 Aug 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/macos-bigsur-kerberos-sso-over-vpn/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;If you&#39;ve read my blogs about &lt;a href=&#34;https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn/&#34;&gt;macOS Catalina Kerberos SSO over Per-App Tunnel&lt;/a&gt; and the &lt;a href=&#34;https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn-followup/&#34;&gt;followup&lt;/a&gt;, you&#39;ll know that this has been a use-case I&#39;m interested in solving.  I put a great deal of effort into filing feedback with Apple and providing steps to replicate the issue.  I was quite excited when I saw the per-app Tunnel improvements specifically mentioned in the WWDC videos, and hoped perhaps some changes were made to enable this functionality.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Using GitHub Actions to Publish Hugo Site From Private to Public Repo</title>
      <link>https://blog.terakedis.dev/post/github-actions-publish-private-hugo-repo-to-public-pages-site/</link>
      <pubDate>Fri, 14 Aug 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/github-actions-publish-private-hugo-repo-to-public-pages-site/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I restarted my blogging journey earlier this year when I started looking into &lt;del&gt;Jekyll&lt;/del&gt; Hugo to generate a static website.  I had past experience with Blogger and Wordpress, but frankly had periodic problems with both platforms that ended up being a time suck.  As it has been, Hugo has been a simplistic publishing method and GitHub a reliable (and FREE) hosting provider.  Yet, my desire to keep my drafts private (.e.g the use of 2 separate repositories) has created a small overhead in that I have to build and manually commit the website changes to the public repository to make them live.&lt;/p&gt;</description>
    </item>
    <item>
      <title>macOS Catalina Kerberos SSO over VPN Followup</title>
      <link>https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn-followup/</link>
      <pubDate>Sun, 22 Mar 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn-followup/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I&#39;ve been going back and forth with Apple about some of the issues I previously found using the &lt;a href=&#34;https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn/&#34;&gt;macOS Catalina Kerberos SSO over Per-App VPN&lt;/a&gt;.  As it turns out, they acknowledged some of the issues I was seeing and are &lt;em&gt;supposedly&lt;/em&gt; working on a fix.  I&#39;ve been watching the past few beta releases for macOS, and I&#39;ve not yet seen anything in the release notes to indicate they&#39;ve implemented any fixes.&lt;/p&gt;</description>
    </item>
    <item>
      <title>New Key Added to SoftwareUpdate Command in macOS</title>
      <link>https://blog.terakedis.dev/post/new-key-added-softwareupdate-command-macos/</link>
      <pubDate>Thu, 19 Mar 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/new-key-added-softwareupdate-command-macos/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;When Big Sur released, I noticed an issue trying to discover what macOS versions were available in SoftwareUpdate.   I worked around this by using the &lt;a href=&#34;https://github.com/munki/macadmin-scripts/blob/main/installinstallmacos.py&#34;&gt;installinstallmacos.py script&lt;/a&gt; to download full installers from the store.  This script/tool is solid, but I was really hoping to be able to just do the updates using softwareupdate.  Awhile back I filed feedback with Apple to the following:&lt;/p&gt;</description>
    </item>
    <item>
      <title>Another GitHub Actions Update - Using Deploy Keys Instead Of Personal Access Tokens</title>
      <link>https://blog.terakedis.dev/post/another-github-actions-update-change-hugo-publish-deploy-keys/</link>
      <pubDate>Fri, 06 Mar 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/another-github-actions-update-change-hugo-publish-deploy-keys/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;When I restarted my blogging journey in 2020, I switched from Jekyll to Hugo hosted in GitHub pages.  It&#39;s been a relatively painless journey, and kudos to GitHub as a rock-solid hosting provider.  I&#39;ve covered it before (&lt;a href=&#34;https://blog.terakedis.dev/post/github-actions-publish-private-hugo-repo-to-public-pages-site/&#34;&gt;Initial Setup&lt;/a&gt; and &lt;a href=&#34;https://blog.terakedis.dev/post/updated-github-actions-publish-private-hugo-repo-to-public-pages-site/&#34;&gt;First Update&lt;/a&gt;), but I&#39;ve been incredibly happy with private-to-public publishing workflow that allows me to keep drafts and work-in-progress hidden.  That said, a recent comment gave me reason to make another update to the Workflow.   Read on for more detail...&lt;/p&gt;</description>
    </item>
    <item>
      <title>My Adapted Digital Bullet Journal via iPad Pro and GoodNotes 5</title>
      <link>https://blog.terakedis.dev/post/organizing-adapted-digital-bullet-journal-ipad-goodnotes/</link>
      <pubDate>Fri, 06 Mar 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/organizing-adapted-digital-bullet-journal-ipad-goodnotes/</guid>
      <description>&lt;p&gt;Last year, my wife introduced me to an entirely foreign-to-me concept of journaling known as &lt;a href=&#34;https://youtu.be/fm15cmYU0IM&#34;&gt;Bullet Journaling&lt;/a&gt;.  For those of you uninitiated, think of bullet journaling as an extensible planner with a heavy focus on the index to help bring order to chaos.  A bullet journal is flexible, allowing you to organize tasks, events, collections (groups of tasks, ideas, and more).   Bullet Journaling was born of a need to bring all the ideas/tasks/events plaguing us in the digital age and put them to paper in a single place.  By doing so, there was less chance of forgetting things, and the repetition of writing helped bring mindfulness to the list.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Deploying Microsoft Defender ATP for macOS using Workspace ONE UEM</title>
      <link>https://blog.terakedis.dev/post/deploying-microsoft-atp-defender-for-mac-using-workspace-one/</link>
      <pubDate>Tue, 18 Feb 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/deploying-microsoft-atp-defender-for-mac-using-workspace-one/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;I had a few folks recently approach me on the MacAdmins slack asking for help with deploying Microsoft Defender ATP for Mac.   We got it working, but it came down to 2 issues:  conflicting documentation and Jamf/inTune specific templates.  Once I was able to parse through the Jamf/InTune documentation, we were able to put together some guidance.  We recently published this guide to the &lt;a href=&#34;https://github.com/vmware-samples/euc-samples/tree/master/macOS-Samples/3rd-Party_Software_Guidance/Microsoft-Defender-ATP&#34;&gt;EUC Samples GitHub Repository&lt;/a&gt;.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Testing macOS Catalina Kerberos SSO Extension Over VPN</title>
      <link>https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn/</link>
      <pubDate>Tue, 04 Feb 2020 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/post/macos-catalina-kerberos-sso-over-vpn/</guid>
      <description>&lt;div class=&#34;notice&#34;&gt;&#xA;  &lt;div class=&#34;notice__label&#34;&gt;Where I&#39;m focused now&lt;/div&gt;&#xA;  &lt;div class=&#34;notice__body&#34;&gt;&#xA;    &lt;p&gt;Heads up - this is an older post. These days my focus is on AI/LLM pipelines, Azure architecture, and automation. If that&#39;s more your speed, here&#39;s some of what I&#39;m working on now:&lt;/p&gt;&#xA;    &lt;ul&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/llms-are-not-magic/&#34;&gt;Why I Stopped Treating LLMs as Magic and Started Treating Them as Components&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/ai-pipeline-reliability-9am/&#34;&gt;Designing an AI Pipeline That Works Everywhere, All the Time&lt;/a&gt;&lt;/li&gt;&#xA;      &lt;li&gt;&lt;a href=&#34;https://blog.terakedis.dev/post/leanwheel-skills-what-i-added/&#34;&gt;leanwheel-skills, a Month In: A Plugin, a Pile of New Skills, and Better Token Math&lt;/a&gt;&lt;/li&gt;&#xA;    &lt;/ul&gt;&#xA;  &lt;/div&gt;&#xA;&lt;/div&gt;&#xA;&#xA;&lt;p&gt;Working at VMware, I&#39;m surrounded by great technology and super-smart folks!  In our portfolio of technologies, the folks in our R&amp;amp;D have recently been putting quite a bit of effort into building out macOS capabilities for our Workspace ONE Tunnel client for macOS.  Workspace ONE admins can leverage the same VMware technology they used to enable per-app VPN for iOS and Android, but now on macOS!  There&#39;s a bit of nuance to configuring the VPN client if you&#39;re previously familiar with iOS (look for my Operational Tutorial soon to hit &lt;a href=&#34;https://techzone.vmware.com&#34;&gt;TechZone&lt;/a&gt;).  That said, the premise is the same -- by configuring the appropriate rules, the Tunnel app redirects traffic from whitelisted applications back into your network through the Unified Access Gateway.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Welcome to Hugo (from Jekyll)</title>
      <link>https://blog.terakedis.dev/2018-06-15-welcome-to-jekyll/</link>
      <pubDate>Fri, 15 Jun 2018 00:00:00 +0000</pubDate>
      <guid>https://blog.terakedis.dev/2018-06-15-welcome-to-jekyll/</guid>
      <description>&lt;p&gt;Welcome to my wholly reworked website!  This time, I&#39;ve dropped the complexity of Wordpress and opted for something significantly simpler:  &lt;del&gt;Jekyll&lt;/del&gt; Hugo and GitHub.&lt;/p&gt;&#xA;&lt;p&gt;Having now hosted the site in GitHub Pages, here was the process I started with Hugo:&lt;/p&gt;&#xA;&lt;ol&gt;&#xA;&lt;li&gt;Created the Github Pages repo (rterakedis.github.io) -- this is where github pages looks for the blog&#39;s generated site files.&lt;/li&gt;&#xA;&lt;li&gt;Created a 2nd Github repo:  rterakedis.github.io.hugo -- this repo holds the source files for Hugo to parse and generate the site.&lt;/li&gt;&#xA;&lt;li&gt;Added the GH Pages repo (rterakedis.github.io) as a submodule for rterakedis.github.io.hugo&lt;/li&gt;&#xA;&lt;li&gt;Edited the config.toml in the hugo files to include the following:&lt;/li&gt;&#xA;&lt;/ol&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;&lt;code&gt;baseURL = &amp;quot;https://terakedis.dev/&amp;quot;&lt;/code&gt;&lt;/li&gt;&#xA;&lt;li&gt;&lt;code&gt;publishdir = &amp;quot;rterakedis.github.io&amp;quot;&lt;/code&gt;&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&lt;ol&gt;&#xA;&lt;li&gt;Add the CNAME file into the rterakedis.github.io repo and enable the custom name/https in the repo settings&lt;/li&gt;&#xA;&lt;li&gt;Ensure the output from &lt;code&gt;hugo&lt;/code&gt; builds into the rterakedis.github.io directory in my local rterakedis.github.io.hugo directory&lt;/li&gt;&#xA;&lt;li&gt;Commit and push everything to GitHub&lt;/li&gt;&#xA;&lt;/ol&gt;&#xA;&lt;p&gt;A few quick links that I found particularly helpful when I was working with Jekyll:&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
