pinouchon

Neurosym labs | Zurich (maybe NY/Boston, maybe SF/Bay area) | ONSITE, INTERNS, VISA | Full-time

I am considering starting an AI/ML research company in the field of neurosymbolic programming (which sits at the intersection of deep learning, program synthesis and probabilistic programming). The goal is to be 100% focused on producing research that is made publicly available, guided by intellectual curiosity and a desire to move the field forward.

I am looking for co-founders or early employees. Although there is an existing company & office that will be ready by the end of summer, not all details are figured out yet. The location is 80% likely to be Zurich (keeping the current office), but could be in the US. The funding is privately secured by other ML-related ventures I was involved with in the past. The name is also a placeholder name and will change. I can give more details privately.

Roles and compensation:

  - Full-time employees: base of around 100-250k USD/annum + equity with standard vesting (negotiable).
  - Interns or residency program: base of around 80-180k USD/annum

(Other research interests: self-play, world models, common sense reasoning, program induction, causality, bayesian inference)

To apply, send me an email to b.crouzier at gmail and including "Neurosym labs application" in the title. Please reach out also if you just have pointers or other things you want to discuss, as I'm still exploring all options available.

Look for Vitalik, balaji srinivasan, naval ravikant podcasts, also chris dixon

Interestingly, the poet with the highest level understanding can be clueless about UTF-8 or what a pixel is

I'm using the qgmlwy layout for a few years now, pretty happy with it. I remapped caps & added a new modifier to have a "special chars" layer and it works pretty well. Overall it's more trouble than it's worth, if I had to redo the whole journey, I would simply learn to touch-type with querty (maybe colemak), but I'm still happy with the end result.

Before that, I had a strong interest in AI & ML, read the Deep Learning book and did the cs231n mooc. So I could train basic neural networks with pytorch but that was it.

We have experience trading since march 2020, so no proper experience in a bear market. We deployed our first algos with nontrivial amounts (> 100usd) the night before the march crash. And yes we did get rekt that night :) Trading strategies work. That it still works is a huge mystery to me. We did beat BTC, we have no long/short bias, and are doing the best during volatile periods (up or down)

Deep Learning (by Aaron Courville, Ian Goodfellow, and Yoshua Bengio) Read cover to cover, then applied these techniques to Crypto Markets (turns out, crypto markets are pretty inneficient)