Everyone talks about AI as if it's a genie in a lamp—a magical black box that just grants wishes. We hear about how it wrote an essay or created a masterpiece, as if we, the humans, are just passive observers. Then you have the other crowd, the ones who say you have to learn every line of code and master every technical assumption to truly get it.
I think both approaches are profoundly wrong.
My philosophy is different. Thanks to a few lucky breaks and student subscriptions, I've had the chance to get hands-on with some of the best models out there. And what I've learned is this: the best way to understand an AI model isn't just to read about it, but to truly use it.
My AI Arsenal
Claude: The Technical Architect
When a creative idea hits, my mind is already in motion. The last thing I want to do is get bogged down in the technical minutiae that follow. That's where Claude comes in.
It's my technical architect, perfectly suited for the detailed, foundational work that needs to be done flawlessly. I give Claude a concept, and it handles the logical structure, the intricate explanations, and all the "how-to" details.
Perplexity: The Instant Researcher
I am, by nature, impatient. My time is for thinking, not for wading through a sea of search results. Perplexity is the solution to that.
When a question pops into my head, I can simply speak it aloud and get a concise summary from multiple sources in seconds. The voice feature allows me to stay in a state of high-level thought while Perplexity handles the research.
Gemini: The Creative Mirror
My most powerful tool isn't something that does work for me; it's something that thinks with me. I use Gemini as a creative mirror.
I'll pitch it a half-formed idea, and it will reflect back a new angle, a clever twist, or a powerful new phrase I hadn't considered. The conversation is a constant refinement process, gathering insights I need to define what the next generation of AI will look like.
Building the Future
This isn't just about using a service. It's about a strategic workflow built around my own unique requirements. I'm not just benefiting from AI; I'm actively figuring out how to build the blueprints for the AI of the future.