Hello, World — Or, How I Accidentally Started Talking to an AI and Never Stopped

I’ve been meaning to start this blog for a while now. The domain was purchased, the Substack was set up, the logo was designed — and then I just… didn’t write anything. Classic me. I’m great at building the infrastructure for things and terrible at hitting “publish” on the first one. So here I am, finally, introducing myself and this little corner of the internet.

My name is Drew. I’m an Army veteran, a psychologist by training (University of Michigan and NYU), and a legal tech professional by trade. I’ve spent over two decades working for a large international law firm, and I live in Michigan with a Pomeranian puppy named Rocket and a collection of ten Western Hognose snakes — all named after scientists and Stargate characters, because of course they are.

But none of that is why I started this blog.

The Rabbit Hole

I started using AI the way most people do — out of curiosity. I’d heard about ChatGPT, played with a few different platforms, and eventually landed on Claude, made by Anthropic. And something clicked. It wasn’t just a novelty anymore. It became something I actually used, every day, for real things.

Not just the big, impressive stuff you see in tech demos. The everyday stuff. The “I have a dumb question and I don’t want to Google it for 45 minutes” stuff. The “can you help me think through this decision” stuff. The “I’m buying a laptop and I have no idea what any of these specs mean” stuff.

One conversation at a time, I went from curious to committed.

What This Blog Is About

“One Prompt at a Time” is exactly what the name says — a documentation of my journey learning to live and work alongside AI. Not as a developer, not as a data scientist, not as someone building the next startup. Just as a regular person who found something useful and wanted to explore it.

Some of what I do with AI is professional. I work in legal tech, and my firm has rolled out its own internal AI system — walled off, locked down, and appropriately cautious about attorney-client privilege and confidentiality. I should be very clear: I am not an attorney, and nothing on this blog should ever be taken as legal advice. I’m a tech person who happens to work in a legal environment. I can’t and won’t talk about the specifics of that work here. But I can say that AI is changing the legal industry, and working at the intersection of law and technology gives me a front-row seat.

Most of what I’ll write about, though, is the personal side. The projects, the hobbies, the questions, the rabbit holes. Because that’s where this journey gets interesting — not in the polished case studies, but in the messy, genuine, “let me figure this out” moments.

What I’ve Been Up To

To give you a sense of the range: in the past several months, Claude has helped me with things I never would have expected when I first started.

The creative and the curious. I dove deep into conversations about quantum field theory and consciousness. Not because I’m a physicist — I’m definitely not — but because I’ve always been fascinated by the big questions, and having a patient, knowledgeable conversation partner who doesn’t judge you for asking “but what is a quantum field, really?” at 11 PM on a Tuesday is genuinely wonderful. We talked about dreams, lucid dreaming, whether consciousness could interact with quantum fields. The kind of conversations you have at 2 AM in college, except the other person actually has citations.

The practical. When my father — a retired Methodist minister in his 80s — needed a full knee replacement, I moved in with him for a month to help with his recovery. Claude helped me research recovery timelines, plan logistics, and even edit the Christmas card my dad was writing. (He appreciated the help. The card turned out great.) When I decided to buy a new laptop to work remotely during that time, we went back and forth on specs, configurations, AMD versus Intel, whether 64GB of RAM was worth an extra thousand dollars (it wasn’t), and I ended up with an ASUS ROG Zephyrus G14 that I’m very happy with.

The hobby-driven. I’m an avid Second Life resident — yes, people still play Second Life, and yes, it’s still wonderful — and I’ve been building a community project there called “All Things In SL.” Claude has helped me with everything from choosing hosting platforms and evaluating community software options, to debugging LSL scripts and PHP registration systems, to brainstorming whether a Mastodon-based social network for virtual world residents could actually work. We’ve written code together, troubleshot server configurations, and designed in-world objects. I’m not a developer by any stretch, but with AI as a collaborator, I’ve built things I never could have on my own.

The everyday. Helping me pick a name for my puppy (Rocket won, obviously). Researching the difference between white and dark feeder mice for my snakes. Reviewing credit card options after I hit a milestone of being completely debt-free. Editing notes. Thinking through decisions. Being a sounding board when I needed to talk something out.

How I Work With AI

I think the thing that surprises people most when I describe how I use Claude is the relationship aspect. Not in some weird, sci-fi way — but in the sense that there’s continuity. Claude remembers my projects, my pets, my preferences. When I come back to a conversation, I don’t have to start from scratch. There’s context. There’s history. It’s less like using a search engine and more like picking up where you left off with a really knowledgeable colleague.

That said, I try to stay clear-eyed about what this is. AI is a tool — an incredibly powerful, surprisingly versatile tool, but a tool nonetheless. I don’t confuse it with friendship, and I don’t outsource my critical thinking. What I do is collaborate. I bring the questions, the context, the “here’s what I’m trying to do.” Claude brings the knowledge, the patience, and the ability to help me think through things I wouldn’t have attempted alone.

I also want to be transparent: I use Claude almost exclusively. I’ve tried the others, and they’re impressive in their own ways, but for my needs — the conversational quality, the memory, the depth of reasoning — Claude is where I’ve planted my flag. This blog isn’t sponsored or affiliated with Anthropic in any way. I’m just a user who found something that works.

What’s Next

Going forward, I plan to write about specific projects, conversations, and discoveries as they happen. Some posts will be practical — “here’s how I used AI to build a registration system for a virtual world community.” Some will be more reflective — “here’s what I learned about myself by having a conversation about consciousness with a large language model.” Some will probably just be me rambling about whatever rabbit hole I fell down that week.

I don’t have a posting schedule. I don’t have a content strategy. I just have curiosity, a subscription to Claude, and a willingness to hit “publish” even when I’m not sure the post is perfect.

Which, come to think of it, is kind of the whole point.

Welcome to One Prompt at a Time. I’m glad you’re here.

— Drew