Open source is your biggest internship
19-year-old me merged a tiny Jenkins Pull request and that commit snowballed into GSoC, the Linux Foundation, the Kubernetes release team, and eventually Meta.
most ppl’s refusal to risk humiliation is their original downfall.
genius is indistinguishable from delusion / being a moron until the outcome arrives.
the funniest part is that when it all works, everyone retrofits the narrative to make it look inevitable.
& when it fails, they call it what they were always going to call it. stupid.
the line between idiocy & insight is fine as hell.
If you're an Arsenal fan and never got the chance to watch Arsenal's trophy lift, you can relive all 44 minutes of Arsenal's title winning celebrations here 😍
“Why argue with an Arsenal fan when you can just wait?” We waited. We won. We are champions of England - and we are just one game away from being crowned champions of Europe. Read my piece on what Arsenal means to me here: https://t.co/J6cg388mRH
It is my understanding, there is a seriously confident expectation now for a decision on the Man City 115 case to be delivered in the coming weeks ( sometime in the summer latest )
All Premier League clubs await official communication.
if you are in college in india in 2026, the market is not dead, it is just filtering harder I was reading an India Skills Report 2026 puts employability at 56.35% which means you need to do something different to stand out and get your foot in the door
one of the easiest health wins is fixing your sleep start time
I’ve tried sleeping 30 minutes earlier for 7 days and it’s starting to make a difference
Did a very different format with @reinerpope – a blackboard lecture where he walks through how frontier LLMs are trained and served.
It's shocking how much you can deduce about what the labs are doing from a handful of equations, public API prices, and some chalk.
It’s a bit technical, but I encourage you to hang in there - it’s really worth it.
There are less than a handful of people who understand the full stack of AI, from chip design to model architecture, as well as Reiner. It was a real delight to learn from him.
Recommend watching this one on YouTube so you can see the chalkboard.
0:00:00 – How batch size affects token cost and speed
0:31:59 – How MoE models are laid out across GPU racks
0:47:02 – How pipeline parallelism spreads model layers across racks
1:03:27 – Why Ilya said, “As we now know, pipelining is not wise.”
1:18:49 – Because of RL, models may be 100x over-trained beyond Chinchilla-optimal
1:32:52 – Deducing long context memory costs from API pricing
2:03:52 – Convergent evolution between neural nets and cryptography
Studying how data moves in the network during inference
The movement of the KV cache across prefill and decode puts some strain on NICs
The world of networking has come up with a great distinction between front end nics and backend nics