[{"body":"","link":"https://www.rockyoliver.com/tags/ai/","section":"tags","tags":null,"title":"AI"},{"body":"","link":"https://www.rockyoliver.com/categories/ai-deep-thoughts/","section":"categories","tags":null,"title":"AI Deep Thoughts"},{"body":"","link":"https://www.rockyoliver.com/categories/","section":"categories","tags":null,"title":"Categories"},{"body":"","link":"https://www.rockyoliver.com/tags/creativity/","section":"tags","tags":null,"title":"Creativity"},{"body":"","link":"https://www.rockyoliver.com/","section":"","tags":null,"title":"Elder IT Geek"},{"body":"","link":"https://www.rockyoliver.com/tags/environment/","section":"tags","tags":null,"title":"Environment"},{"body":"","link":"https://www.rockyoliver.com/tags/ethics/","section":"tags","tags":null,"title":"Ethics"},{"body":"","link":"https://www.rockyoliver.com/post/","section":"post","tags":null,"title":"Posts"},{"body":"","link":"https://www.rockyoliver.com/tags/","section":"tags","tags":null,"title":"Tags"},{"body":"Let me preface this by saying I don't have the answers; in fact, I honestly believe there isn't a \u0026quot;right\u0026quot; answer.\nBut I want to take a moment to explain my approach to the controversial, disruptive technology known as AI.\nI want to make sure that you understand that this is my own personal position, based on the people in my life, my career, and my own feelings. And here's the thing - I'm not even comfortable with where I'm at on this. I don't know that there is a comfortable place to be when it comes to AI. And in all likelihood I'll change my mind - often - as this new landscape unfolds.\nBut here's my reality.\nSetting the stage First, I'm an Elder IT Geek (hence this blog). I'm married to a wonderful person who's also a professional actress, director, and educator. My cousin Ward Jenkins is an award winning illustrator and animator. I've seen, firsthand, the ugly side of AI - using it to steal the likeness, voice, art style, content, and livelihood of people close to me. And in no uncertain terms, that is wrong. My wife has had to be much more careful when considering audition requests - she has to carefully read the brief and the contract, because on multiple occasions there's been a clause in there that gives the producers the right to use your image, voice, etc. for \u0026quot;generated content\u0026quot; in perpetuity - with no compensation. Basically, they're actor farming so they can generate AI versions of the actors that sign that contract.\nAnd we all know the controversy with other creatives, especially artists, musicians, and writers. Their work was used, without their permission or compensation, to train pretty much every AI and LLM out there.\nAnd that's wrong too.\nBUT (you knew that \u0026quot;but\u0026quot; was coming...)\nIn what I do, in IT, AI has been revolutionary. Like, I cannot express how revolutionary. Productivity has skyrocketed, especially for people who know how to use AI as a productivity tool, a force multiplier in development, and not let it run amok unchecked.\nAnd to be brutally honest, if you are a developer and you are NOT learning about AI, and learning how to use it effectively, you WILL lose your job to someone who does.\nAnd that's been a very hard pill to swallow.\nHere comes the elephant... At the advent of the Industrial Age and the internal combustion engine, there was an immediate explosion in manufacturing productivity - but it came at a cost. Blue collar workers were treated almost like slaves. The environment was being ravaged by all the smog and pollution, and no one at the time understood long term environmental impacts - or if they did they didn't care because they were getting rich. If you were a common worker, you didn't get a piece of the pie - you were the mincemeat filling it.\nAnd this is always what happens when a disruptive technology comes along that essentially reshapes society. Technology always outpaces regulation and legislation, and every time the people who are in line to gain the most barrel forward headlong before legislation catches up. And local governments are happy to let it happen, because every politician loves tax revenue not tied to a voter.\nDuring the Industrial Revolution regulations did catch up - but it took decades. And that's unfortunate, but it's a pattern that repeats throughout history, especially in the 20th and 21st centuries.\nAnd we're seeing the same pattern again with AI and their data centers. Data centers are gobbling up agricultural land and draining aquifers dry. A single large data center can consume millions of gallons of water per day for cooling. And local officials are letting it happen because they got a sweetheart deal and \u0026quot;can't say no\u0026quot;.\nYet the Industrial Revolution soldiered on. And so will the AI Revolution, for better or worse.\nBefore you go looking for stones... The internet, and more specifically social media, has given rise to the voices of many moral absolutists. It's very easy to voice your opinion - loudly, passionately - today. Even I do it, more than I care to admit. But things are rarely black and white, if ever.\nSo before you start looking for stones to throw, I just want to caution you to remember that we all live in glass houses. We are all, collectively as a society, guilty of living in and perpetuating a knowingly destructive industry - the petroleum industry. We are, for better or worse, a global petroleum economy. It literally touches every single facet of every single person's life. And if it suddenly went away tomorrow, humanity wouldn't survive. And the worst part? We don't have a backup, and we're running out of it.\nAnd this is what makes times like these so difficult. You can both be opposed to something and still have to be a part of it. During the Industrial Revolution workers kept working while fighting for their rights. People kept driving cars while fighting for emissions controls. Workers continued to work for substandard wages until unions and the minimum wage were put into place.\nYou can hold two opposing views, and both be true.\nThe uncomfortable reality In the end, there is no \u0026quot;one right answer\u0026quot;. Right now, especially in the early stages of the Age of AI, the greedy people are going to exploit the gaps and margins at the expense of society and our environment. And to be brutally honest, we won't be able to stop it, because quite frankly there's too much money to be made - and humans have proven time and again to be shallow, short-sighted and self-centered.\nSo what can we do? What can you do?\nI think each of us has to carefully consider how we respond to the AI Revolution, and make the best, difficult choices that we can live with. Each of us has to balance our principles with the reality we're living in, which means that we may have to compromise those convictions to survive. Then all that remains is living with your decisions.\nPersonally, I have accepted my uncomfortable reality. I must use AI in my job, or I won't have a job - or a career. That's just the way it is. However that doesn't mean I'm a full-throated proponent of AI. I will actively defend the rights of creatives like Ward and my wife. I will continue to write my Congresspeople to encourage them to put regulations in place to protect our environment and our people. In the meantime, all I can do is continually reassess the situation, adjust my actions accordingly, and hope that I can continue to look myself in the mirror, however uncomfortable that may be.\nI want to leave you with my favorite quote from my favorite sci-fi novel series, Dune:\nRun faster. History is a constant race between invention and catastrophe. Education helps, but it's never enough. You must also RUN. -- Leto Atreides II (God Emperor of Dune, Frank Herbert)\n","link":"https://www.rockyoliver.com/post/elephant-in-the-room---article-03/","section":"post","tags":["AI","ethics","environment","creativity"],"title":"The AI Elephant in the Room"},{"body":" I've spent over 35 years in IT doing almost everything there is to do — engineer, architect, team lead, trainer, manager, executive, author, speaker, and entrepreneur. I've worked for international consultancies, pre-IPO startups, and even ran my own shop a couple of times.\nI've shipped code, led teams, written books, and stood on stages in front of thousands of developers all over the world.\nRight now I'm a Senior Architect working in R\u0026amp;D for a Really Big Company, where I get to do what I love most — dig into hard problems and figure out what's actually true versus what's just noise. And I teach others how to do the same.\nI've been around long enough to watch \u0026quot;transformative technologies\u0026quot; come and go. I know the difference between genuine signal and expensive marketing.\nContainers? Real.\nRust? Real.\nKnowledge Management? Ask me sometime — I'll buy the drinks. (We'll be very drunk by the end.)\nAI is real. And it's the most interesting moment of my career.\nI'm not here to tell you AI is going to save the world or end it. I'm here to work the problem — with 35 years of scar tissue guiding the way.\nThat's what this blog is about.\nThe Name \u0026quot;Elder IT Geek\u0026quot; isn't self-deprecation. It's a category. If you've been in this industry long enough to remember when \u0026quot;the cloud\u0026quot; was just a PowerPoint slide with a fluffy shape on it, you're one of us.\nWe've got context. We've got pattern recognition. We've got scar tissue.\nThose things matter. Especially now.\n","link":"https://www.rockyoliver.com/about/","section":"","tags":null,"title":"About"},{"body":"","link":"https://www.rockyoliver.com/categories/ai-best-practices/","section":"categories","tags":null,"title":"AI Best Practices"},{"body":"","link":"https://www.rockyoliver.com/tags/development/","section":"tags","tags":null,"title":"Development"},{"body":"","link":"https://www.rockyoliver.com/tags/it/","section":"tags","tags":null,"title":"IT"},{"body":"So you've decided to byte (ha! see what I did there?) the AI bullet; but you've seen the horror stories about how people have used AI to write elegantly bloated and broken code. How do you avoid creating crappy code at breakneck speed? By understanding what AI is - and isn't. Once you understand the nature of the beast, you can go about taming it, ultimately turning it from a wild stallion into a racehorse.\nThis is the first article in a series that will help you understand the nature of AI from a coder's perspective, and give you some practical advice for not only getting started, but getting the most out of your AI coding experience. Now we're not going into expert mode - we're going to talk about some simple, practical tips and tricks that will help you become more productive than you've ever been in your life.\nLet's begin by understanding what AI actually is...\nSo, what is this AI thing, anyway? The vast array of AIs out there - ChatGPT, Claude, Copilot, etc. - are all based on what's known as a Large Language Model, or LLM for short. Basically an LLM is an incredibly complex algorithm known as a neural network (I won't get in the weeds here, you can look it up - it's a deep rabbit hole) that's been trained on vast amounts of data - that's the \u0026quot;Large\u0026quot; part. The \u0026quot;Language Model\u0026quot; part is the thing that makes AI seem almost human - anthropomorphic, if you like fancy words. It's coded to both understand and produce information in a natural language way.\nAnd while there are a myriad of ways to use an AI - and some of them, OK a LOT of them, are not so honorable - we're going to focus on using AI in IT, specifically in development. So while some of the things we're going to be talking about over this article series may apply to areas outside of development, the focus of everything here is from an IT and coder's perspective.\nIt's best to have a clear picture of what AI is, and isn't, clearly defined in your head. As you begin to understand the nature of AI you can begin to harness its power.\nPicture a room... I find the best way to think about AI is to picture a room full of energetic, young university CS graduates. And these aren't just any graduates - they are familiar with damn near every programming language ever released, and are really good at it from a tactical perspective.\nBut they have zero common sense, and zero experience in the real world.\nThey can churn out code in record time - and that code might even work, locally, on one machine.\nBut they failed to take into consideration user load, concurrency, network limitations, throughput, latency, and the other myriad things that go into producing a production level application.\nThe worst part? They will confidently, enthusiastically tell you that they're done and it will work.\nAnd they are categorically wrong.\nYou see, AI is great at the immediate, tactical work of slinging code. But it's horrible at seeing the big picture.\nAnd that's where you come in. You have to be the grizzled, seasoned adult in the room. You have to be the one who diligently spot checks their work, challenges their output, and holds them to task. Because you are responsible for what they produce.\nOne other thing - AI has a horrible memory. Like goldfish level horrible. Each session is like a fresh room of CS graduates. Sure, they may have heard rumors in the hallway about what the group was doing, but they have zero context of what was actually said or done. No details, no decisions, no specifics. So you have to develop a strategy for managing your context across sessions.\nWe'll talk more about that in the next article.\nLet's talk strategy... Now that you're beginning to understand the nature of AI as a development tool, let's start talking about some general strategies that you can use to keep AI focused and productive. Here's some hard earned advice.\nManage your AI development like you're working with a room of enthusiastically ignorant programming experts You need to not be afraid to challenge everything your AI tells you, no matter how confident it is - and it will be really, really confident. Keep in mind that no matter how much it sounds like a human, it's not. You won't hurt its feelings. Hold it accountable. Remember, you're responsible for what it puts out.\nIt's up to you to keep AI from veering off-course As your AI is working, actually read what it's doing. AI has the habit of \u0026quot;hallucinating\u0026quot;, or making up things as it goes along; and then it will continue working off that hallucination at 100MPH (that's 162KPH for the rest of the world), and if you're not paying attention it will have built something ridiculously huge, and mind-numbingly wrong. It will also forget things you just told it - and make the same mistakes you told it not to make before. It's up to you to make sure your AI stays cogent and focused.\nWork in short sessions There's this concept in AI of session context. The best way to think about this is that AI is keeping mental notes and pointers about everything you're doing. But just like when you watch a four hour movie, after awhile you may forget exactly what happened at the beginning. AI is the same. Therefore it's better to work in smaller, more focused and concise sessions. Break large efforts into smaller discrete tasks, done across multiple sessions - usually one for every task or two. Next article I'll give you some strategies for maintaining context across sessions.\nMake AI check itself Remember, AI isn't a human, and doesn't have feelings or an ego - so don't be afraid to challenge an AI to check itself. Quite often I'll tell it to double check what it just did, just to make sure it's right. Sometimes I'll give it questions to ask to help focus the review effort, and other times just a general review will turn up things that are wrong, things that were forgotten, or things that could have been done better. Hell, it even often finds its own bugs. You can even have a different LLM review code just written - for instance you could do a coding session in Claude, and then have ChatGPT or Gemini review that code independently like a digital peer review of the code. This mirrors what effective development groups do already (and if you're not getting your code peer reviewed before you deploy it - shame on you).\nDon't let the tail wag the dog This is a big one. Most of the AI development horror stories you hear are caused by developers not actively monitoring or managing the output from AI. Whether it's from ignorance, laziness, or oversight, allowing AI to develop something without oversight and review is almost guaranteed to generate something big, beautiful, and unusable.\nMake sure it knows, emphatically, that you don't want a \u0026quot;yes man\u0026quot; Remember that most AIs are designed to please their users. And while this might work great for the average human, this is not what we want as IT geeks. So you need to tell it, explicitly, that you want and expect input, recommendations, advice, and pushback when warranted. And tell it to remember that as a standing rule from now on.\nStarting to get the picture? Hopefully this little introduction is starting to paint the picture of what an AI actually is and does when used in development. It's an incredibly powerful tool - but only when used responsibly and correctly.\nIn future articles in this series we'll dive into some practical best practices for getting the most out of your AI coding efforts. These articles won't be like the plethora of digital pablum out there with overengineered strategies or veiled marketing attempts pitching a completely unnecessary tool. These are going to be tips and strategies learned by myself and others that will help you get the most out of your AI today.\nSee you next time.\n","link":"https://www.rockyoliver.com/post/what-ai-is---article-02/","section":"post","tags":["AI","IT","development"],"title":"What AI Is - And Isn't"},{"body":"","link":"https://www.rockyoliver.com/categories/ai-and-your-career/","section":"categories","tags":null,"title":"AI and Your Career"},{"body":"So, you’re staring down the barrel of AI - and it’s squarely pointed at you. If you’re an IT professional, you may be dubious. Tentative. Excited. Even scared.\nAnd rightly so.\nHow we, as IT professionals, understand and adopt this technology will determine not only our success in the near term, but the trajectory of our entire career overall. But the way you work with AI (and whether you want to or not, you WILL be working with AI) will greatly determine how your career unfolds.\nWho you are, what you do, and where you are in your career has a direct impact on how you should approach AI. To keep this article from turning into a novella, I’m focusing on three IT roles - Elder IT Geeks, Newbie IT Weenies, and C-Level executives and Managers. Let’s start with the one I’m most familiar with, Elder IT Geeks.\nElder IT Geeks You’ve been doing this IT thing for decades. Maybe you’ve got some gray hairs - or no hair - and you’ve just been cruising along, teaching the newbies, fighting the fires, building the 401K.\nAnd then AI happens.\nNow for most of us, the immediate response is disdain. There’s NO WAY some overzealous search engine can do your job, right? And it’s just another thing for me to learn and forget, when the fad fades.\nWell, you’re both right, and wrong.\nFirst off, where you’re wrong - AI is not going away. There’s too much money being invested in it, and it’s already permeating damn near every facet of our lives, especially those of us working in IT. So, you need to make your peace with it.\nThe good news is where you’re right. It can’t do what you do, at least not right now. You’ve got decades of knowledge, experience, and scars that no AI can replicate. And all those lessons you learned and knowledge you gained is your AI superpower. This is your time to be a freaking gray haired rock star.\nYou should picture AI as a room of hundreds of eager young developers who just graduated from college, have impressive knowledge of a myriad of programming languages and platforms, but have zero real world experience. And they REALLY want to please you - but their enthusiasm can cause them to confidently make absolutely wrong choices. They have the technical knowledge, just not the real world experience of taking the “big picture” into consideration.\nAnd that’s where you become the rock star. You DO have decades of experience, and are able to see the big picture in a way that no AI can (at least for now). And because you can keep them on the right course, correcting them as they go, you will instantly go from just an “elder geek that’s forgotten more than he knows” to one of the most productive people on your team.\nBut what if you’re towards the beginning of your career?\nNewbie IT Weenies Well, I was trying to think of a way to soften this, but there just isn’t one. If you’re, say, fresh out of college up to in the middle of your career, where you’re still a lunch pail developer honing your craft, then really the only advice I can offer is:\nBe scared.\nAI was built to basically do your job, and do it much much faster than you can. So what does that mean for you? The good thing is that not all hope is lost; but you’re going to need to be very proactive in what you do to “future proof” your career in IT.\nThe first thing you need to do is start learning the stuff that’s not just the Xs and Os. Learn the deeper, weird stuff about architecture, scalability, concurrency, corporate politics, etc. It’s basically the skills we used to use to discern between a software developer and a software consultant. These are the skills that will keep you ahead of AI, because AI just isn’t good at that secondary and tertiary level of thinking and planning.\nAlso, you need to be an AI-fu master. Learn all you can about it - different LLM models and what they are good (and bad) at, building MCPs and skills, adding AI to products, understanding agentic coding, etc. Instead of viewing AI as an adversary, turn it into an ally.\nOh, and I recommend buddying up to those Elder IT Geeks before they retire. Take them out to lunch, buy them a couple of beers, and listen to their war stories about difficult projects. You can greatly speed up your move up the IT ladder by growing your experience through their experience.\nIf this section resonated with you, here’s even more motivation for why you want to grow your career past worker bee developer - your C-Level executive or manager may not listen to what comes next.\nC-Level And Management You’ve seen all the hype about how AI can revolutionize your business and grow your bottom line. You might have even dipped your toe into the AI waters, and seen immediate results. And that’s a good thing - for now. But make sure you’re not just looking at your feet, but also at the road ahead.\nThe mistake I most often see executives and managers make is only being focused on the bottom line. This is a short-sighted strategy, and pretty much always leads to Bad Things Happening. Just like Elder IT Geeks look past the immediate task towards secondary and tertiary concerns, you as an executive need to look past short term profitability and not lose sight of the things that have a profound impact on your long term business health - especially when they can’t be measured in dollars and cents.\nAnd here’s the real land mine. Right now, you may have noticed a marked increase in productivity from your IT Department, especially from those Elder IT Geeks - now you know why. But here’s where keeping your eyes down the road comes into play. Do not fall into the trap of thinking “well, my gray haired geeks are much more productive than my worker bee developers. I can just fire all of them, and instantly become more profitable!” That’s staring at your feet, and not looking ahead.\nBecause in a few years those Elder IT Geeks are going to retire:\nWho’s going to take their place?\nIn the Southern US we have an old saying, “make sure you don’t eat your seed corn”, meaning make sure you don’t cannibalize your future to fill your belly in the present. Your “seed corn” in this instance are the Newbie IT Weenies that followed my earlier advice - and they’ll stand out from the ones blindly plodding along. And if you’re not sure who they are, ask the Elder IT Geeks, because they’ll have already spotted the promising young stars that are paying attention and making moves to improve themselves.\nIn the end there will be a happy medium - you probably will be able to reduce headcount somewhat, and remove those folks who are not adapting to the changing landscape. Incidentally, this happens every time a disruptive technology gains prominence - some people adapt, some people thrive, and some people fall away.\nBut make sure you plan for the future. Work with your Elder IT Geeks to proactively train up the promising Newbie IT Weenies. Because in the end, how AI benefits your organization long term is highly dependent on what you do in the short term to plan for sustained productivity. Stare at your feet and not look ahead, and you’ll be asking yourself “What happened” as your profitability suddenly plummets and you’re left wondering why.\nYour Future is in Your Hands No matter what your role is in IT - Elder IT Geek, Newbie IT Weenie, or C-Level Executive, you can either watch the world change around you because of AI - and possibly get left behind - or you can embrace it and plan for your future. All it takes is a little planning and foresight - but you’ve got to be the one who looks ahead.\nRemember the old saying - “Success is where preparation and opportunity meet.” Well, the opportunity is here; the preparation is completely within your hands.\n","link":"https://www.rockyoliver.com/post/ai-and-your-career---article-01/","section":"post","tags":["AI","IT","career"],"title":"AI May Not Be The End Of Your World, But It Could Be The Beginning Of Your Future"},{"body":"","link":"https://www.rockyoliver.com/tags/career/","section":"tags","tags":null,"title":"Career"},{"body":"","link":"https://www.rockyoliver.com/series/","section":"series","tags":null,"title":"Series"}]