yauneyz

talysman.app does the same thing as Brick but for desktop. So you can code/etc with a physical layer of protection

Definitely a game changer, maybe even more than Brick because the desktop is where I do all my work

I think the biggest hurdle remaining is that all these landmark results are generally counter-examples.

Proving something in the affirmative often requires the creation of an entire new sub-field of math, or new tools. Think of Fermat's Last Theorem or something like that.

These results, while impressive, are clever constructions using existing techniques. It isn't clear that AIs can build new machinery like this. But if/when they can, yeah it is probably game over.

I had the exact same problem with LeechBlock override, so I built talysman.app, which is basically Brick (the phone app) but for desktop. If your problem is solely on your phone, just use Brick, same principle but on the phone.

The key point is that you can only unblock when a physical usb drive (paired with the app) is inserted.

I've found this is the "just right" amount of friction. Too much friction (like impossible to override) and you have to be careful about when you turn it on, in case you genuinely need Reddit or YouTube or something. Too little friction and you just always override. But having to walk over to the next room to override is perfect.

I had it write a short story about Vader and Palpatine discovering the Graham Schmidt process. It wasn't the greatest thing ever but it got the mood right and understood what Graham Schmidt was. It was crazy at the time

I'm sure astronomy solves much harder "needle in a haystack" problems and we could probably use their methods

Genuinely curious - did you use a coding agent for most of this or does this level if performance take hand written code?

I think part of the problem is a temptation to believe that we can have out cake and eat it too.

If the people on charge of deciding when to use the cameras were morally perfect, we have all the upside and none of the downside.

The problem is we live in a fallen world and that will simply never work.

Nevertheless it is a siren song that causes us to repeatedly make the wrong trade

Something fluffer. Sankey diagrams will never live that down lol

Depends how you configure it. If you like things like tiling window managers and keyboard driven computing, Linux is in a category of its own.

Location: Palo Alto, CA, USA Remote: Yes Willing to relocate: No Technologies: Python, PyTorch, Numpy/Pandas/sk-learn, SQL, JS, React, Clojure, ClojureScript, NodeJS Resume: https://docs.google.com/document/d/1Innj32msgL2BBg6NH-gEzGAa... Email: zac [dot] yauney [at] gmail [dot] com Interesting Project: https://www.thinky.dev

When I was young I wanted to be a theoretical physicist, but then I learned that computers let you use math to model and impact the "real" world, and I was turned towards data science. Masters in Applied Math, experience with Python and its data science stack and deep learning.

Full stack JS experience as well, using React. I also have built a fairly sophisticated Electron app using ClojureScript and what is essentially React.

Looking for roles in data science or otherwise general software engineering

Check out thinky.dev - it is primarily focused on 2 right now, but I think it is really good at it.

It was also built because of academia, my masters thesis

If this is Electron, try pdf.js - really good rendering, you can create a text layer (for text selection, etc). Probably the best result per effort you can get

Location: Palo Alto, CA

Remote: Yes

Willing to relocate: No

Technologies: Python data stack, SQL (w/ dbt), Pytorch, MERN stack, intermediate Clojure, some Rust, Linux

Resume: https://docs.google.com/document/d/1jy37umFhfek96_-MAwSgxJYG...

Email: In CV

I'm actually part of a research team working on using a wrist mounted spectrometer to actually measure glucose continuously and non-invasively. Turns out it's a really really hard problem.

My professor has talked about this. He thinks that the real gem of the deep learning revolution is the ability to take the derivative of arbitrary code and use that to optimize. Deep learning is just one application of that, but there are tons more.