R. Gabriel Esteves

We need philosophy because no one was ever radicalized into revolutionary politics from learning mechanical engineering. Technical skills will only make the system more efficient, but we always need at least the possibility of reimagining the system entirely.

" Todos los libros pueden dividirse en dos clases: libros del momento y libros de todo momento". John Ruskin

What’s exciting about programming? The joy programming can bring. That heart racing moment you built something amazing and then you hit the go button and it worked and it did something new. Sometimes we forget what got us into coding in the first place and we suck the fun out.

Marijuana must be legalized. All our people imprisoned because of it must be returned home to their families and their records expunged.

Disinfo nerds! As we close out 2019, let’s marvel at amazing women that are leading research, analysis, collaboration, investigations in the disinformation space. @camillefrancois & I have often reflected on this privately, but now there’s a list to follow https://twitter.com/i/lists/1211816787467264001 …

"But Holmes," said Watson, astonished. "Don't you know that the Earth orbits the sun?" "What the deuce is it to me? You say that we go round the sun. If we went round the moon it would not make a pennyworth of difference to me or to my work." —RE: Same Code for CPUs/GPUs/TPUs

The hardest work is to decide what to work on and who to work with. Not the work’s itself.

Princeton History PhD candidate Xiyue Wang kept working on his dissertation and learned two languages while in an Iranian prison so I really don't want to hear about it from the rest of you.

"To me programming is more than an important practical art. It is also a gigantic undertaking in the foundations of knowledge." - Grace Hopper

An API has two sides: first, the mental models underlying it, which should straightforwardly map to the mental models of domain experts. Then, the actual objects/signatures of the API. API design is mostly about designing the 1st. The 2nd is just a mechanical consequence of it.

I've been thinking about the software stack for machine learning. Tools I'd love to see. 1. Pip for pretrained models. 2. Version control for datasets. 3. GPU-friendly CI. Travis CI, Circe CI don't support GPUs. Jenkins is a pain. 4. Fast dataframes. Why is Pandas so slow?

The best way to learn more about a system is to try to instrument it.

Signs you need a break: * loosing interesting in things you’re usually passionate about and bring you joy * short temper and easy to provoke * taking everything personally * constant overthinking * emotionally overwhelmed * always exhausted/tired Take care friends. Take a break.

In many cases the best place where to find a clear and self contained explanation of a topic/experimental technique, is a chapter in the thesis of someone who spent their PhD working on it. So why don't we publish those chapters in the "journal of clear explanations"?

"Debuggers don't remove bugs. They only show them in slow motion." - Unknown