alexispomares Goto Github PK

Update Feb '24: For the past couple of years I've been working largely on closed-source proprietary software — under a separate account — in a startup that demanded most of my time, until very recently. Now, I'm excited that this 2024 I will be publishing some open-source code in here again — starting with a few recent promising, fun experiments in Deep Reinforcement Learning!

This is Alexis. My goal is to blur the frontiers between mind and machine, using Deep Learning and Brain-Computer Interfaces.

Some interesting repos from my journey so far:

• Deep Learning classification of EEG responses to multi-dimensional Transcranial Electrical Stimulation, a.k.a. DL-EEG-TES.
• A simple auditory Oddball Paradigm for EEG/Neurostimulation experimental sessions.
• Deep Reinforcement Learning to play both chess and Clash Royale at superhuman level, based on DreamerV3.
• A full-body 3D-printed phantom for medical training and research in image-guided surgery.
• A way to play mobile games using your (ECG) heart rate in real time.
• A simple setup to have an Aleph Zero decentralized validator node running in under an hour.
• Samantha OS (pt 1 and pt 2), a WIP to eventually have a human-level LLM use my computer for me, based on OpenInterpreter.
• Testing the current limits for LLMs to accurately simulate human behavior, based on Generative Agents.

. . .

Other links about me:

• Built Lymbic.ai, a startup to develop neuroadaptive apps that integrate the user's mental states (e.g. mood) in real time.
• Feel free to reach out via LinkedIn!

. . .

FUN FACT: I once 3D-printed a 1:1 life-size replica of my own brain (from an MRI scan for research). Check out this beauty!

Step 1: Scan Yourself 🤯	Step 2: Extract your Brain 🧠	Step 3: Use as Paperweight! 🤓