Monolith MMI (mind-machine interfaces) is a collective of 5 inventors, determined to create technology that provides more precise and accurate data about the state of the human mind. We want to improve the accuracy of measuring information communicated by the brain by creating technology that is generalizable, precise, and useful. Our current goal is to create an Electroencephalogram that can accurately capture strings of thoughts, without relying on physical movement, or EMG signals. We plan to do this by manufacturing our own PCB, researching additional forms of synaptic activity capture, and developing a machine learning model to generalize to diverse use-cases and users.
Our team is called Monolith BCI (on Twitter), and we are creating a novel PCB design and ML model to process electrical signals generated by large clumps of neurons in the brain with an EEG. Henry is working on the ML model (Bramble), as well as a bunch of parts of the project, Cheru and JC designed the PCB with 0 PCB experience before this project. I’m creating a research paper, and dataset (FLUX- the Framework for Learning and Understanding Cortex Activity) to help fine-tune our LLM model.
We want to create an (up to 32 channel) PCB and an ML model that decode electrical signals in the brain that pick up patterns that humans are not able to do using data analysis and filtering with only fast Fourier transforms and power spectral decomposition.
We plan to get to a point where we can can accurately capture strings of thoughts, without relying on physical movement, or EMG signals.
By the end of our 3rd sprint goal is to have our own PCB working with our ML model to play Tetris and Pong without using EMG signals (we’ve already successfully used four modalities- jaw clench, blinking, focus, and non-focus), but all of these rely on EMG signals, and our product will only be viable once we’re able to capture thoughts alone- but we believe we’re really close to getting there.
This funding will be used for compute, PCB manufacturing, electrode purchase (and potentially manufacturing), an Airbnb for the team to stay in SF during our third sprint, food and transport.
Cheru (16), Henry (15), JC (17), me (19), and Nila (18). Everyone but me has worked at Hack Club and is proficient at coding, creating projects, and generally hacking (we all love solving problems, using flipper zeroes, talking about rationality and what we don't like about it, understanding physics, ML, and math (to varying degrees), and reading up on deep tech projects). Henry is a 2023 Atlas Fellow, Nila and I are on gap years, JC is working full time at hack club and Cheru works part time, and is on his last year of high school (he condensed sr. and jr. years of high school into one). None of us have ever worked on a project of this scale, but we've only had two sprints are already have a working PCB (Trillium v2) that can collect electrical data from the brain that integrates with our rudimentary fine-tuned ML model.
This project could fail due to poor generalization of the ML model to brain signals, and a lack of training data specific to our model. The outcome of failure would be a waste of money, but we've open sourced and shared every part of our project thus far, so people would know exactly where we got stuck and what we wished we would have done differently.
$300 -> Hack Club OnBoard grant
$1000 -> 1517 Medici Grant
$500 -> Yush via Hack Grants
App status across various funders