Hackathons are no longer the Computer Science meta.

Sun, 02 Aug 2026 01:00:00 +0530

There was a time when winning a hackathon required something vaguely resembling computer science. You needed to know how your application worked, how your backend communicated with your frontend, why your database wasn’t on fire, and occasionally, what the hell a WebSocket was.

Those were simpler times.

Today, you need three things: a decent prompt, a collection of AI subscriptions that would make your bank account file a police complaint, and enough design sense to make a generated landing page look like it belongs in a YC pitch deck.

Sorry, we're out of tomorrow's RAM.

Wed, 15 Jul 2026 10:00:00 +0530

There are two kinds of people who preorder things before they exist: gamers and trillion-dollar AI companies. The former get a collector’s edition with a plastic figurine. The latter get priority access to HBM wafers that haven’t even finished pretending to be silicon.

If you’ve been wondering why GPUs cost roughly the GDP of a small island nation, the answer isn’t just NVIDIA’s leather jackets. It’s memory.

Modern AI training is no longer compute-bound in the traditional sense. You can keep adding more TFLOPS, tensor cores, and exotic matrix instructions, but eventually every accelerator reaches the same existential crisis: waiting for data. A GPU without memory bandwidth is just an expensive space heater.

Journal on Ujaan Mukherjee

Hackathons are no longer the Computer Science meta.

Sorry, we're out of tomorrow's RAM.