A year ago, I was writing code the old way — Googling error messages, copy-pasting Stack Overflow answers, and spending half my time on boilerplate I didn't want to write. Today? I have an AI assistant that runs 24/7 on my server, coding agents that build entire projects for me, and local language models running on my Mac Mini. AI hasn't just entered my daily life — it's running it.
The Starting Point: Claude Code
Claude Code was my first real taste of what AI coding agents could do. I'd describe what I wanted, and it would scaffold projects, write functions, debug issues, and iterate on my feedback. It wasn't perfect — sometimes it'd go down rabbit holes or make questionable architectural decisions — but the productivity boost was undeniable.
I started using it for everything: web projects, automation scripts, data processing tools. The moment I realised I could describe a problem in plain English and get working code back in seconds, I knew things had fundamentally changed.
Not a Replacement, an Amplifier
Here's the thing people get wrong about AI coding tools — they think it's about replacing developers. It's not. It's about amplifying what you can build. I'm a nurse by training, not a software engineer. Without AI, I'd never have been able to build half the things I've built. Claude Code didn't make me a coder; it made me a builder.
Then Came OpenClaw
If Claude Code was a power tool, OpenClaw is an entire workshop. OpenClaw is a personal AI assistant platform that runs on your own server and connects to your messaging apps, calendar, email, and pretty much everything else. Think of it as having a hyper-competent assistant who never sleeps and actually understands context.
I set it up on my Unraid server, connected it to Telegram, and suddenly I had an assistant that could manage my schedule, search the web, control devices, and even help me write blog posts (yes, including this one — the irony isn't lost on me).
What Makes It Different
The difference between OpenClaw and ChatGPT-style assistants is context and agency. ChatGPT is a conversation. OpenClaw is a system — it has access to my files, my tools, my services, and it can take action on my behalf. It remembers things between conversations, it learns my preferences, and it proactively suggests things I might need.
It runs on my hardware, with my data, under my control. No sending everything to a third-party cloud. That matters to me.
Local LLMs on the Mac Mini
The latest addition to my AI setup is running local language models on my Mac Mini. There's something deeply satisfying about having a capable AI model running entirely on your own hardware, no internet connection required.
The Mac Mini is surprisingly capable for this. With enough RAM and a decent processor, you can run models that would have required a GPU cluster a couple of years ago. I've been experimenting with different models for different tasks — smaller, faster models for quick queries and code completion, larger models for more complex reasoning tasks.
Why Bother With Local?
Speed, privacy, and reliability. Local models respond instantly with zero network latency. Your data never leaves your machine. And you're not subject to API rate limits, service outages, or usage costs. It's the difference between renting a tool and owning one.
The trade-off is capability — local models aren't as powerful as the frontier cloud models yet. But they're getting closer every month, and for many day-to-day tasks, they're already good enough.
The AI Stack That Runs My Life
Here's what a typical day looks like:
• Morning: OpenClaw sends me a digest with emails, calendar events, weather, and news
• During the day: Claude Code handles any coding tasks I throw at it
• Afternoon: Local LLMs on the Mac Mini help with quick research, writing, and brainstorming
• Evening: OpenClaw manages tasks, sets reminders, and keeps things organised
• Throughout: AI-assisted photo editing, content creation, and workflow automation
None of this was planned. It evolved organically as I found tools that worked and integrated them into my life. The result is a personal AI ecosystem that saves me hours every week and lets me build things that would have been impossible a year ago.
The Possibilities Really Are Endless
That's not hyperbole — it's the honest reality of where we are right now. The combination of cloud AI, local models, and agent platforms means the barrier to building useful, personalised tools has never been lower.
Want to automate your home? There's a model for that. Want to build a web app? Describe it and iterate. Want an assistant that knows your schedule, your preferences, and your workflow? Set it up once and it works forever.
I'm not an AI researcher or a tech giant. I'm a nurse from the Philippines living in Australia who likes tinkering. If I can build this kind of AI setup, anyone can. The tools are here, they're affordable, and they're getting better every single week.
What's Next
I'm exploring more ways to integrate AI into my workflows — both professional and personal. The clinic venture I'm working on? AI is going to be a core part of that. The blog you're reading? Hosted on a CMS that was set up with AI assistance.
The question isn't whether AI will change how we work and live. It already has. The question is whether you'll use it or ignore it. I chose to lean in, and I haven't looked back.
If you're on the fence about diving into AI tools, start small. Pick one thing — coding, writing, automation — and find a tool that helps. You'll be surprised how quickly it becomes indispensable.
No comments yet