How small changes made AI feel more like a thinking partner than a tool.
Intro
My usage of AI has gone up drastically in 1 year. I spend more time on ChatGPT + Gemini + Perplexity + Claude together, than any other website or segment.
But over the last few days, I have noticed that I am learning to use it better.
These aren’t big, technical breakthroughs. They’re small, day-to-day shifts that have changed how I interact with AI
Here are my top 3 learnings from the last month — simple tweaks that made a big difference in how AI fits into my work and life.
1. Smarter Prompts — The 3-Layer Formula
Earlier, I used to type prompts like I’d write a Google search — short, quick, and direct.
That worked okay, but the replies often felt generic.
Over time, I started structuring my prompts in three parts: Context, Task, and Tone.
Something like:
“Context : I am 28 years old. My salary is 80,000 per month. I live in Noida. You are a personal finance expert helping me plan my monthly budget.
Task: Suggest a spending breakdown that balances quality of life & savings.
Tone: Practical and realistic, not preachy. Give rationale behind your approach”
It’s simple, but that one change made a huge difference.
The responses felt more thoughtful and closer to what I actually meant.
Whether I tried it on ChatGPT or Claude, the difference was clear — the AI seemed to “get” my intent better.
Takeaway: Treat prompts like briefs, not searches.
2. Top-Down Approach for Bigger Tasks
When I started using AI for larger things — like a monthly marketing plan or blog structure — I used to dump everything into one long prompt.
The results were usually messy and hard to fix.
Now I take a top-down approach.
I start with the basics — title, structure, sections.
Then I move to subpoints.
And only after that do I ask for detailed content.
It’s like building a skeleton before adding the flesh.
That small shift changed everything. The process feels calmer and more deliberate.
Instead of rewriting confused outputs, I refine them layer by layer.
The final result is sharper — and I have better understanding of how I reached it.
Old approach:
“Make a complete 30-day marketing plan for an online t-shirt store.”
Result:
Big wall of text. Confusing order. Hard to tweak without breaking everything.
New approach:
Step 1: “List 5–6 key marketing areas for an online t-shirt store.”
Step 2: “Create a week-wise outline for those areas.”
Step 3: “Now, fill each week with specific actions and sample captions.”
Result:
Layered structure. Easy to modify. Each stage builds logically into the next.
(Interestingly, this mirrors how AI itself works best — when tasks are broken into smaller, structured steps.)
Takeaway: Structure first. Content later.
3. AI for Personal Finance & Travel Planning
We take a 15-day trip almost every quarter.
And in the last two, I’ve leaned on AI for everything — budgeting, itinerary, hotels, the works.
I usually write a short, 5-6 line brief: where we’re going, how many days, and what kind of travellers we are.
Within seconds, it gives me a full plan — estimated costs, travel routes, even food suggestions.
What really surprised me was the accuracy.
The final spend was within about 5% of what AI had predicted.
That’s not just luck — that’s a hint of genuine intelligence.
It’s not perfect, of course. Some suggestions are too generic, or miss the local feel.
But it gives a solid starting point — something you can tweak and make your own.
And that’s the part I love: the heavy lifting is done, so you get to focus on the fun parts.
Bonus Tip — Cross-Check Everything
A small practice that’s helped a lot — run the same query through another AI tool.
I usually cycle between ChatGPT, Gemini, Perplexity, and Claude.
It’s like asking for a second (or third) opinion.
Each model has a slightly different strength.
Sometimes one catches what another misses.
Takeaway: Two AIs are better than one.
Closing
Looking back, these three learnings — smarter prompts, top-down thinking, and practical use — all point to the same thing.
AI hasn’t just made me faster; it’s made me think more clearly.
The better I brief it, the better it briefs me back.
And that’s the real shift — it’s less about AI getting smarter, and more about me learning how to think with it.
I would love to know how you use AI tools.