I pick hard problems I don't know how to solve yet. Not because the world necessarily needs them solved. Because the moment it finally works — that "oh shit, I actually did it" — that's the whole point.
That's why I built a self-play system where one AI breaks what another builds. Why I'm implementing a research paper from scratch instead of using someone else's code. Why I built a structural engineering tool having never touched HTML Canvas or Shapely before.
The complexity is the point. The not-knowing is the point.
BTech AI · Amity Kolkata · 2027
