Cole Foster | Machine Learning Engineer

Hi, I'm Cole Foster.

A

Self-driven, quick starter, passionate programmer with a curious mind who enjoys solving a complex and challenging real-world problems.

About

I am a student at the University of Miami majoring in Computer Science, Mathematics, and Neuroscience. I enjoy problem-solving and coding, and strive to bring 100% to the work I do. I have worked on technologies like Python, PyTorch, C, CUDA, C++, MySQL, PostgreSQL, Git, Github, Docker, Kubernetes, R, Java, and MATLAB during my bachelor's. I have nearly 3 years of professional research experience which helped me strengthen my competency in Python, PyTorch, machine learning, and software engineering principles. I am passionate about developing complex applications that solve real-world problems impacting millions of users.

Languages: Python, C, CUDA, C++, R, Java, MATLAB, Bash
Databases: MySQL, PostgreSQL, SQLite
Python Libraries: NumPy, Pandas, OpenCV, Matplotlib, Seaborn, Pillow
Machine Learning Frameworks: PyTorch, Scikit-Learn, Hugging Face, XGBoost, PyTorch Geometric
Tools & Technologies: Git, FastAPI, Docker, Kubernetes, CMake

Looking for an opportunity to work in a challenging position combining my skills in Machine Learning and Software Engineering which provides professional development, interesting experiences and personal growth.

Experience

Wuchty Lab

Machine Learning Researcher

Developed a scalable pipeline for graph embedding and learning tasks, processing >8M heterogeneous relationships
Achieved 98% lower loss on link prediction and clustering tasks using attention with graph neural networks
Accelerated code execution by 415x through vectorization and GPU optimization
Tools: Python, PyTorch, PyTorch Geometric, Hugging Face, Git, Bash

November 2023 - Present | Miami, Florida

DSaCB Lab

Machine Learning Researcher

Designed a Visual Transformer (ViT) evaluation pipeline processing nearly 500M attention tokens
Engineered meta features for transformer evaluation, filtering >99% of non-interpretable features
Maintained vision model metrics (ACC, AUC, F1) with a >95% reduction in training data using ensemble learning
Tools: Python, PyTorch, Hugging Face, Git, Bash

May 2023 - Present | Miami, Florida

Ali Lab

Data Scientist

Invented a feature selection algorithm which removed nearly all experimental noise, improving measurable treatment effect by 3 orders of magnitude
Constructed a configurable data processing pipeline, decreasing outside experimental variance by >90%
Led automated image processing initiatives that improved data collection speeds by 242x per sample
Tools: MATLAB, Git

June 2022 - July 2024 | Miami, Florida

Projects

Models From Scratch

From linear regression to transformers in Python, C, CUDA, and C++

Accomplishments

Languages: Python, C, CUDA, C++
Tools: Git, CMake, FastAPI, Docker, Kubernetes
Loads data and fits/predicts with a multitude of machine learning models quicker than standard Python libraries
Followed good software engineering practices including organization and documentation
Implemented a CI/CD pipeline with GitHub Actions, Docker, and Kubernetes to ensure continued delivery and updates
Developed an intimate understanding of the math behind machine learning models