
2025 was a year of output and motion. I [shipped](/projects) a lot, [traveled](https://instagram.com/_a_gryffindor) a lot, and learned fast. It was also a year where I got clearer about what *doesn't* work for me - especially when it comes to focus and the kinds of commitments I allow into my life.

### What I'm proud of

**1) I kept shipping, even when the ideas changed.**
I launched a bunch of small things and experiments - some quick, some slow - and I'm glad I didn't stop building just because everything wasn't perfectly aligned. A few examples:

* I shipped a prompts library ([Promptsmint](https://promptsmint.com)) - 100+ Daily Users - and continued improving it as a real asset.
* I launched a couple of micro-builds (small tools, mini-products) that helped me practice distribution + iteration.
* I shipped an iOS app that took longer than expected, but I stuck through the slower "finish it" part.

**2) My AI + product instincts leveled up.**
This year I got materially better at building AI/agent-like workflows and turning them into interfaces that feel usable. Less "demo energy," more "someone could actually use this daily." That shift matters.

**3) I had a lot of real-world fun (and it wasn't a distraction).**
I did Mumbai/Goa, a long Bali trip, and a month in Thailand/Vietnam, plus a few shorter trips. See more on [Instagram](https://instagram.com/_a_gryffindor). Travel didn't magically solve anything, but it consistently reset my brain and gave me energy. That's valid.

**4) Reading came back.**
I restarted reading and it stuck. It gave me a calmer baseline and better thinking. Quietly one of the highest ROI habits I picked up.

### What didn't work (and I'm done pretending it will)

**1) Fragmentation kills me.**
I can handle intensity. I don't handle *too many parallel streams* well. When projects + plans + people all demand attention at once, my output drops and my mental state follows.

**2) Some relationships/commitments were a net negative.**
Not naming specifics, but here's the rule I earned the hard way:
**If a relationship isn't net positive - energy, stability, and support - it doesn't deserve a long trial period.**
I don't mean "perfect." I mean: if it repeatedly costs focus and creates stress without adding something real back, it's a no. I've learned to exit earlier instead of negotiating with my own future.

**3) Activity ≠ progress.**
I did a lot. But the gap between "busy" and "moving toward the goal" is real. Some of my confusion came from winning small battles while the bigger direction stayed fuzzy.

### The patterns I noticed (the honest ones)

* **When I start my day reactive, I lose the day.**
* **When I compound one thing weekly, things move fast.**
* **When my boundaries are weak, my focus becomes a shared resource - and I lose.**
* **When I build things that are user-facing and practical, I'm in my zone.**

### What I'm taking into 2026

I'm keeping the plan simple:

* **Fewer bets, more compounding.** One main asset, weekly progress, no constant resets.
* **More public artifacts.** Small, frequent proof of work beats occasional big announcements.
* **Protect focus like it's a job requirement.** Deep work blocks are not optional.
* **Health + reading stay steady.** They make everything else easier.

If you want to do your own yearly review, I wrote the exact framework I use here: [The Yearly Review (Simple + Actually Useful)](/blogs/yearly-review).



Hey, I'm
Aman

AI Product Engineer

My 2025: shipping, signal, and a few uncomfortable rules

Hey, I'm Aman

AI Product Engineer

My 2025: shipping, signal, and a few uncomfortable rules

Hey, I'm
Aman