pls trust I look better irl 😭
Jonathan Xu
451 posts
noninvasive neural decoding @alljoined
San Francisco, CA
Joined July 2016
- Bless the immigration gods, I got my second O-1 visa! Approval in 3 business days 🤩 Great work @lighthousehq_
- Some platforms can be used in ways not intended. Here's @azlenelza and I doing a high-five after winning a game of Codenames on @figmadesign
GIF - Introducing Alljoined (@_alljoined), my new thing! By applying deep learning to large-scale EEG datasets, we aim to decode the entirety of the conscious experience non-invasively. Proud to be backed by Khosla Ventures, with participation from Conviction, Compound, Artis, Xtal,New episode of Core Memory out today
00:00 - dropping a BCI dataset: Announcing Alljoined1, a dataset made specifically for EEG-image reconstruction. We got accepted to CVPR's DCAMI workshop, see you in Seattle! arxiv.org/abs/2404.05553 We've seen remarkable progress in decoding images from the brain through work that
- no sir nothing to declarepacking for sf, I think I have everything I need
- Come work on mind reading at @_alljoined We're now state of the art on EEG -> image, and some really crazy stuff in the pipeline. So excited to see this world-class team come together If you're a really good MLE, SWE or Neuroscientist - check us out!
- Excited to share a really cool project: reconstructing video from fMRI brain scans using ViTs with temporal attention, CLIP, and stable diffusion! 🧠➡️📺 Building with the team has been awesome mind-video.com
GIF - Excited to share our latest work on image reconstruction! Announcing MindEye2: Shared-Subject Models Enable fMRI-To-Image with 1 Hour of Data medarc.ai/mindeye2
- we just got accepted to ICML and ICLR 💪 💪Excited to share our latest work on image reconstruction! Announcing MindEye2: Shared-Subject Models Enable fMRI-To-Image with 1 Hour of Data medarc.ai/mindeye2
















