I'm Ken Reid — a Senior Data Scientist at Rocket Mortgage with a Ph.D. in Artificial Intelligence from the University of Stirling. My research spans evolutionary computation, optimization, and machine learning, with publications at top venues including GECCO and IEEE SSCI. I design and build data-driven solutions across Python, Java, JavaScript, and more — from automated testing frameworks to large-scale optimization systems.
Peer-reviewed research papers and preprints
This thesis addresses challenging real-world employee rostering and shift scheduling problems using state-of-the-art metaheuristic techniques. Novel approaches including Variable Neighbourhood Search, Evolutionary Ruin & Stochastic Recreate, and hybrid matheuristic methods combining metaheuristics with Integer Programming were developed and evaluated against real-world data provided by BT. The research demonstrates how computational optimization can solve complex, highly-constrained scheduling problems that arise in large-scale workforce management.
Try these data-driven web apps — no setup required
Paste your Steam profile and get your gaming life roasted. Gaming personality, head-to-head comparisons, achievement stats, account value estimation, genre breakdown, and downloadable summary cards.
Upload your GoodReads export and discover your reading personality, genre breakdown, rating analysis, fun page facts, and a quotes collection tool with themed PDF/HTML/Markdown export.
Compare Avalanche vs Snowball debt repayment strategies with interactive charts. Balance transfer analysis, sensitivity modeling, bi-weekly payments, and customizable summary cards.
Upload your Letterboxd data export and get absolutely roasted. Film personality types, data-driven roast commentary, multi-user comparison, contrarian scores, and shareable summary cards.
Data science & research projects on GitHub
Factorio game interface for optimizing belt balancer layouts using evolutionary computation. Companion code for the GECCO '21 paper.
Variational Autoencoder for drug discovery — generating novel molecular structures using deep generative models.
Convolutional Neural Network for pneumonia detection from chest X-ray images — a deep learning tutorial for medical imaging.
Generalized NLP pipeline — topic modeling, sentiment analysis, and text classification using modern natural language processing techniques.
Interactive Google Colab demonstration of Simulated Annealing — reconstructing a target string with visualized convergence, temperature decay, and acceptance probability.
Hands-on machine learning tutorials in Google Colab — covering classification, regression, unsupervised learning, MLPs, CNNs, and RNNs.
Python simulator for comparing debt payoff strategies — avalanche vs snowball methods with financial data analysis and visualizations.
Data pipeline using R, SQL Server, and Tableau to analyze long-term global temperature trends and climate patterns.
Presented through Michigan State University's BEACON Center for the Study of Evolution in Action. This talk accompanies the paper "The Factory Must Grow: Automation in Factorio" and has been viewed over 21,000 times.
Technical write-ups and project deep-dives. View all posts →
GitHub
Google Scholar
LinkedIn
ORCID iD
ResearchGate