If there’s one myth I wish someone shattered earlier, it’s this:
You don’t have to do everything alone.
Throughout the early stages, I mostly kept to myself.
I thought “real researchers” figured things out solo, spending long nights battling code until breakthroughs happened.
It was lonely. It was draining.
Then, I started talking — really talking — to my research teammates.
Not formal meetings. Not forced check-ins.
Just genuine conversations about the confusion, the doubts, the “I have no idea what’s happening” moments.
One of those conversations changed everything.
I was rambling about trying to implement a complex transformer variant for segmentation, hoping it would magically solve my efficiency issues.
One teammate listened patiently, then said something simple:
“Why are you trying to make it complicated? Sometimes polishing a simple method is harder and more impactful.”
That hit me.
I realized I was chasing complexity , thinking complexity would equal innovation.
But refining simplicity — now that was true skill.
After that, I revisited simple ideas like shift operations .
Instead of adding more, I started asking:
- Can I make it lighter?
- Can I make it faster without killing performance?
- Can I make it more robust for messy, real-world inputs?
We brainstormed together. We challenged each other’s assumptions.
And over time, I stopped feeling like I was wandering aimlessly.
I had a direction.
I had a purpose.
The “click” wasn’t a single moment. It was a series of small realizations , built through discussion, reflection, and humility.