Howdy visitor, it's a pleasure to see you here. I’m Alice Williams, a technologist and trans woman from Texas, now based in Washington State. Welcome to my corner of GitHub, where I document and share work not under private licensing.

My work focuses on interpretable machine learning, visual knowledge discovery, and human–AI collaborative systems, with the vision of building AI that thinks visibly and reasons transparently to assist human judgment rather than replace it, while sharing this work openly and learning from others.

This is a living document of my work and provides more context than my formal CV.

Research finished after undergraduate studies:

5. Lossless Visual Analysis of Multidimensional Data in Collocated Paired Coordinates for Machine Learning
Accepted for presentation at HCII 2026 conference Authors: Alice Williams & Boris Kovalerchuk
Presents Collocated Paired Coordinates and paired axial plane arrangements for high-dimensional data visualization and simplified Hyperblock (HB) visualization.

4. Quantifying AI Model Trust as a Model Sureness Measure by Bidirectional Active Learning & Visual Knowledge Discovery
Published in MDPI Electronics journal in the Special Issue Women's Special Issue Series: Artificial Intelligence
Authors: Alice Williams & Boris Kovalerchuk
Proposes Model Sureness to measure trust as a function of reliability for ML model prediction accuracy on given data using iterative supervised learning and visual knowledge discovery.

• Demonstrated that 20–80% of training cases (≈50% on average) could be removed without degrading performance.
• Tools: IterativeSurenessTester

Research began and completed during undergraduate studies:

3. High-Dimensional Data Classification in Concentric Coordinates
Presented at IEEE IV 2025 conference
Authors: Alice Williams & Boris Kovalerchuk
Introduces Concentric Coordinates (CoC) — a lossless generalization of parallel coordinates for interpretable high-dimensional classification.

• Proposes the Generalized Iterative Classifier (GIC)
• Supports tunable rotation, scaling, and axis order
• Tools: Java_Tabular_Vis_Toolkit

2. Boosting Classification Models with Human-in-the-Loop VKD
Presented at HCII 2025 conference; published in Springer’s Lecture Notes in Artificial Intelligence (LNAI) series
Authors: Alice Williams & Boris Kovalerchuk
Presents the CIVL framework combining visual analysis and computational methods to improve classifier performance and explainability.

• Matches AdaBoost accuracy with far fewer parameters
• Tools: Java_Tabular_Vis_Toolkit, HyperblockParser, InLineCoordinatesCoefficientSolver

1. Synthetic Data Generation in Circular Coordinates
Presented at IEEE IV 2024 conference, published in IEEE conference proceedings, awarded a Best Paper Award
Authors: Alice Williams & Boris Kovalerchuk
Proposes static and dynamic circular coordinate systems for labeled synthetic data generation.

• Outperforms SMOTE across 14 classifiers
• Tools: Dynamic_Coordinates_Vis_System

Ordered from most recent to earliest.

My software has run on websites, servers, and personal systems for businesses, individuals, and classrooms since 2015.
View my professional CV here.

• Graduated from Plano Senior High School (2011), completing two semesters of AP Computer Science under a former mainframe programmer and participating in three programming competitions.
• Over ten years of professional experience in system administration, web development, full-stack software engineering, data automation, infrastructure management, and computer science and mathematics education (2015–present).
• Founder, consultant, and independent technologist serving individuals, small businesses, startups, and research labs.
• Earned a Bachelor of Science in Computer Science and Applied Mathematics from Central Washington University; graduated Magna Cum Laude (2025) with significant undergraduate research, conference, and competition experience.
• Research Assistant at the Visual Knowledge Discovery and Imaging Lab (CWU), publishing three peer-reviewed papers under Dr. Boris Kovalerchuk, collaborating with lab members, and remaining an active GitHub maintainer (2023–2025).
• Teaching Assistant for Computer Architecture I and II (300-level courses), Algorithm Analysis (400-level course), and Mathematical Computing (300-level course).
• Admitted M.S. student in Computer Science at Western Washington University; member of the Jag Lab under Dr. Filip Jagodzinski (2025–present; on hiatus for health).
• Experience includes AI EdTech product design, ML consulting, interpretable machine learning research, web systems development, and Linux server administration for distributed services.
• Lifelong technologist, engineer, programmer, scientist, and artist.

• Visualization for Computation & Explanation — Visual structures for computation, model refinement, and interactive analysis.
• Visual Knowledge Discovery (VKD) — Extracting structure from high-dimensional data through interactive visual methods.
• Visual & Interpretable Machine Learning — Designing inherently interpretable models grounded in traceability and reasoning.
• Human-Expert–Assistive AI Systems — Augmenting expert reasoning through hybrid visual, symbolic, and statistical approaches.
• Comparative AI Analysis — Evaluating explainable, classical, and black-box methods across robustness and trust metrics.
• Data Mining & Multidimensional Visualization — Preserving geometric structure while extracting insight from complex data.
• Automated & Semi-Automated Decision-Making — Structured human-in-the-loop reasoning systems.
• Human–AI Interactive Mathematics & Scientific Discovery — Collaborative theorem development, hypothesis generation, and assistive research automation.
• Human–Computer Interaction (HCI) — Designing interfaces for interpretable, user-guided AI systems.

Note: Areas listed in no particular order.

"I believe in intuition and inspiration. Imagination is more important than knowledge. For knowledge is limited, whereas imagination embraces the entire world, stimulating progress, giving birth to evolution. It is, strictly speaking, a real factor in scientific research.”
— Albert Einstein, 1929

Quoted by Dr. Boris Kovalerchuk and Dmytro Dovhalets in: Visual Knowledge Discovery and Machine Learning from 2018 on Chapter 7 Interactive Visual Classification, Clustering and Dimension Reduction with GLC-L on page 173.

"It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience."
Quoted by many as: "Everything should be made as simple as possible, but not simpler."
— Albert Einstein, 1933

I’ve been writing code since before middle school and began studying it seriously in 6th grade (2004–2005).
I write code to learn, explore, experiment, build tools, create art, earn, and bring ideas to life.
My goal is to design simple, understandable systems and theory with the utmost reproducibility, modularity, and usability.

Open to collaboration, research, or new opportunities.
Coming soon, Calendly meeting setup automation, my inbox is difficult to keep up with.
📬 alice.williams.tech@gmail.com