About me

Hi there! I am Manyi Yim.

Driven by an enduring curiosity about free will, human intelligence, and the intricacies of brain function, my early academic pursuits focused on biological neural networks. I developed mathematical and computational models to address neuroscience questions through rigorous analysis of neural data. My journey has since evolved, with recent years bringing a profound fascination with artificial neural networks (ANNs), especially their remarkable capabilities in natural language processing and computer vision. I am keen to pursue cutting-edge research in these areas to understand the inner workings of ANNs, developing innovative and interdisciplinary approaches to complex problem-solving.

Here I would like to share with you some of the projects I have been working on. You can reach me at manyi.yim@gmail.com.

Ongoing Projects

Project	Codes	Packages / References
A Python interface on the client end to run Ollama models on a server	Python
Correlation Structure in the Weights of Transformers across LLMs	Python	Correlation, Linear Algebra

Machine Learning / AI Projects

Project	Codes	Packages / References
REST API with Docker to serve Ollama models on Hugging Face Spaces	Docker + Bash	Docker + Bash + Ollama
Gemma 3 Multimodal Model	Python	Ollama + Streamlit
Inference with Google Paligemma 2 Vision-Language Model (VLM)	Python	PyTorch, transformers, streamlit
Facial Recognition using FaceNet-PyTorch: Can my face be recognized?	Python	PyTorch, scikit-learn, FaceNet
Facial Tracking using FaceNet-PyTorch: Can my face be tracked?	Python	PyTorch, FaceNet
Sentiment Analysis using nltk	Python	NLTK, gradient descent, logistic regression
Sentiment Analysis using a Tensorflow bi-directional RNN	Python	Tensorflow, LSTM, GloVe embeddings
Sentiment Analysis using BERT	Python	Tensorflow, transformers, BERT

Selected Modeling & Data Analysis Projects

Project	Codes	Math/Concept Explained
Place-cell capacity and volatility with grid-like inputs (eLife)	Python	SVM, Perceptron, Multichoose
Transforming the Choice Outcome to an Action Plan in Monkey Lateral Prefrontal Cortex: A Neural Circuit Model (Neuron)	Python / MATLAB	DBSCAN (clustering)
Intrinsic rescaling of granule cells restores pattern separation ability of a dentate gyrus network model during epileptic hyperexcitability (Hippocampus / Springer)	NEURON / MATLAB	-
Impact of intrinsic biophysical diversity on the activity of spiking neurons (PRE)	Python	Numerical Integration, SDEs to Fokker-Planck Equation

Scientific Research Project

Project	Codes	Packages / References
Analysis of grid cell activity in different environments	MATLAB	Image Processing Toolbox, Statistics and Machine Learning Toolbox, Signal Processing Toolbox