I<\/span>‘ve spent most of my career in financial services, with a sideline as a freelance writer. More recently, I became a children’s book author, inspired by my grandchildren. But a software developer? That was never on my radar.<\/p>\n In fact, if you had asked me a year ago to write computer code or build a mobile application, I would have chuckled. While I’ve always been interested in technology and innovation, I never quite had the time or energy to learn programming.<\/p>\n My perspective on what I could achieve changed dramatically with the late-2022 arrival of ChatGPT. Captivated and intrigued, I quickly realized how large language models (LLMs) fundamentally altered the landscape of learning and creation.<\/p>\n I dove into experimentation, peppering the models with questions and requests: Help me plan a five-day itinerary through Santa Fe’s historical sites. Tell me which plants grow best where I live. Explain how LLMs work\u2014like I am 12. Walk me through building a basic calculator application.<\/p>\n The responses came back clear, patient, and judgment-free. I realized that the new technology afforded me unexpected capabilities. Even when I struggled to grasp a concept, AI could explain it to me, step by step and from various perspectives, until it clicked.<\/p>\n I began with practical software projects to test what was possible. My first breakthrough was creating an app that tracked what I’d watched on TV and helped me discover new shows. Simple enough for a programmer, perhaps, but thrilling for me. Next came a cash flow calculator to help manage short-term finances.<\/p>\n I learned some important lessons from these early efforts. For one thing, AI doesn’t think like we do. When I told it to make a button blue, it wouldn\u2019t necessarily know I meant the same shade I\u2019d used elsewhere. If I asked it to add a particular feature, it might not understand my rationale unless I told it why.<\/p>\n It was like learning a new language\u2014not the kind used in programming applications like Python and JavaScript, but the occasionally counterintuitive and precise communication style that often works best with an artificial mind.<\/p>\n I relied on techniques, many acquired through trial and error, to make these interactions more productive. Giving the LLMs defined personas helped significantly: “You are an app programming and security expert.” I formatted my queries\u2014”prompts” in AI speak\u2014with headings and bullets that clearly conveyed what I wanted to see.”<\/p>\n What seemed challenging at first\u2014confusing jargon, unfamiliar concepts, and the meticulous approach needed to solve thorny problems\u2014gradually became manageable. The barriers weren’t about capability; they were about familiarity.<\/p>\n Early on, I adopted a “committee” approach, working with different models to leverage their unique strengths. Sometimes, I’d have ChatGPT draft the initial code, ask Claude to critique the logic, and then query Gemini about other ways to make things work. It became a collaborative dance between human creativity and artificial intelligence, with me as the conductor.<\/p>\n I’d show one model what another had produced and ask, “What’s wrong with this approach? How could it be made better? What did it miss?” When it came to creating software, this proved especially useful for security and efficiency. The LLMs would catch each other’s oversights, suggest optimizations, and help me understand why one strategy was better than another.<\/p>\n The process wasn’t without friction. AI models occasionally “hallucinate,” inventing facts that don’t exist. Their knowledge base isn’t always up to date. When I was building my TV tracker, I had to learn about terms like \u201cAPIs\u201d and structured data\u2014concepts I was only vaguely knowledgeable about.<\/p>\n Sometimes the hardest part was discovering what a particular model didn’t seem to know or couldn\u2019t figure out. But I eventually realized\u2014after bouts of grumbles and bangs on the keyboard\u2014that when I couldn’t resolve an issue after repeated attempts, another AI colleague could step in and make things right.<\/p>\n Despite the occasional frustrations, I believe my age gave me an advantage throughout the journey: patience. At 67, I wasn’t in a hurry. When things didn’t quite work out as I wanted, I persisted. I engaged in relentless testing\u2014creating prototypes, saving my work, making sure everything functioned, and then moving on to the next iteration.<\/p>\n With each project, my confidence grew. Eventually, I decided to tackle something more ambitious: an app that addressed the frustration I had experienced finding discounts for seniors like me. That project ultimately became the foundation for a simple, senior-friendly app I built to help older adults easily locate discounts\u2014an idea born from my own experience and a practical way to apply the skills I had gained. Unimpressed by what was out there, I built something better. AI supported me at every stage.<\/p>\n Creating something out of nothing was thrilling, for sure. But the real revelation wasn’t that I had managed to come up with a polished product I could be proud of. It was the confidence that I could do it again once I put my mind to it.<\/p>\n Many years ago, I had an idea for a children’s video game. I pitched it to firms that specialized in creating these applications, and in a few cases, they expressed genuine interest\u2014but only if I had already done the work of putting the pieces together. Back then, I saw it as a herculean undertaking based on what I knew and the time I had available.<\/p>\n Now, though, I believe I can bring it to life, this playful vision in my head. And the more I think about it, the more it looks like something I’m going to try.<\/p>\n That’s the transformation I didn’t anticipate. Not just learning to turn ideas into tangible reality but discovering that doors I thought were seemingly shut had quietly opened. The impossible had become merely challenging, and the difficult had become achievable.<\/p>\n This journey has taught me far more than the basics of creating computer applications. It renewed my appreciation for curiosity-driven lifelong learning. The barriers that stopped us from acquiring new skills a decade ago\u2014the need for expensive courses, access to experts, an extensive time commitment\u2014have fundamentally changed.<\/p>\n I’m not suggesting that everyone go out and build an app. But whether your next chapter involves writing, volunteering, learning a foreign language, or coding, you now have powerful tools at your fingertips that will listen, support, and teach you with infinite patience.<\/p>\n In truth, the obstacles that hold us back are often more psychological than real. You don’t need to be an expert to start. You just need curiosity\u2014and perhaps a dash of impatience with the way things are.<\/p>\n 67-year-old discovers coding through AI<\/p>\n","protected":false},"author":3,"featured_media":16887,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[211,213],"tags":[],"class_list":["post-16884","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-winter-2026","category-winter-2026-columns"],"acf":[],"_links":{"self":[{"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/posts\/16884","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/comments?post=16884"}],"version-history":[{"count":4,"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/posts\/16884\/revisions"}],"predecessor-version":[{"id":17066,"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/posts\/16884\/revisions\/17066"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/media\/16887"}],"wp:attachment":[{"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/media?parent=16884"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/categories?post=16884"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/healthyaging.net\/magazine\/wp-json\/wp\/v2\/tags?post=16884"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}When AI Became My Teacher<\/h3>\n
Learning to Speak Machine<\/h3>\n
A Collaborative Dance<\/h3>\n
The Gift of Patience<\/h3>\n
A Dream Revisited<\/h3>\n
The Lesson: It’s Never Too Late<\/h3>\n
Michael Panzner is a freelance writer, children’s book author, and the creator of Seniors Save Money<\/a>, a free app available on the web and the Apple App Store.<\/a><\/h5>\n","protected":false},"excerpt":{"rendered":"