← cd ~
shane_cairns_resume.sh

$ whoami

Shane Cairns

email: syre@duck.comlocation: St. Louis, MOlinkedin: www.linkedin.com/in/shane-c-b5b948135/

$ cat summary.txt

Ph.D. candidate in Computer Science (GPA 3.75) specializing in adversarial machine learning, Adaptive Resonance Theory (ART), and robust incremental learning. First author on novel adversarial robustness research achieving 89-100% white-box attack success rates and state-of-the-art defensive training methods. Led DARPA-funded CV pipeline integrating ART with YOLO detection. Hands-on HPC experience (SLURM, multi-GPU PyTorch, CUDA). Strong software engineering practices.

$ cat education.json

Ph.D., Computer Science

Missouri University of Science and Technology | Rolla, MO | Expected May 2027

GPA: 3.75/4.0

Honors: Kummer Innovation & Entrepreneurship Fellow

Research supported by DARPA L2M Program

B.S., Computer Science

Missouri University of Science and Technology | Rolla, MO | Dec 2022

GPA: 3.7/4.0 (Major)

Honors: Distinguished Scholar Award, Dean's List

Graduated in 3.5 years

// Relevant coursework

Machine LearningAdvanced Topics in AIClustering AlgorithmsEvolutionary ComputingProbabilityAlgorithmsMarkov Decision ProcessesCloud & Big DataComputational Bayesian Methods

$ cat experience/research.json

Graduate Researcher - Adversarial ML & Adaptive Resonance Theory

Missouri S&T, Applied Computational Intelligence Lab (ACIL) & KICAIAS | Aug 2023 - Present

# Adversarial Robustness of Incremental Learners (First-Author Research)

  • Developed WB-Softmax, a novel differentiable attack objective for non-differentiable prototype-based models, achieving 89-100% white-box attack success across USPS, MNIST, and Fashion-MNIST
  • Designed progressive two-stage selective adversarial training that achieved best-in-class robustness (AURAC: 28.2%, 64.5%, 41.3% on respective benchmarks)
  • Introduced separation-aware training using incremental cluster validity indices (iCVIs) for geometry-based diagnostics, improving high-perturbation robustness by 4x on USPS
  • Conducted rigorous evaluation with anti-gradient-masking sanity checks following RobustBench best practices

# DARPA Computer Vision Pipeline

  • Integrated ART classification modules into YOLO object detection models; benchmarked accuracy/latency trade-offs across YOLO variants
  • Authored modular PyTorch codebase to swap classifier heads and visualize feature-map activations
  • Built config-driven training/evaluation pipelines with automated checkpointing and artifact logging

# HPC & Infrastructure

  • Executed large-scale experiments on SLURM clusters with multi-GPU PyTorch; optimized data pipelines for I/O-bound workloads
  • Implemented CI with lint/format hooks; maintained unit/integration tests for data loaders and attack modules

$ cat experience/industry.json

Software Engineer Intern

Ford Motor Company | Remote | May 2022 - Aug 2022

  • Modernized legacy flat-file systems by implementing SQL datastore with Java API for secure loan data access
  • Delivered Qlik Sense dashboards for mainframe datasets, improving visibility for Ford Credit stakeholders

Software Developer Intern

Howmet Aerospace | Cleveland, OH | May 2021 - Aug 2021

  • Improved shipping traceability across North American flow path; maintained 10+ production ASP.NET applications

$ ls publications/

"Robustness of Fuzzy ARTMAP to Adversarial Attacks and Progressive Adversarial Training for Streaming Learning"

Cairns, S., Brito da Silva, L.E., Petrenko, S., Wunsch, D.C., & Liu, J.

[2025] Under review

"ART-GAN: Fuzzy ARTMAP as a Prototype-Based Discriminator for Generative Models"

Cairns, S., et al.

[2026] In preparation

$ find ./projects -type f

Sport Analytics Tool

Nov 2022 - May 2025

  • Developed CV routines to detect events from match film (shot attempts, passes, heat maps) with explainable coach insights
  • Implemented feature extraction and model-based classifiers; exported visual overlays and reports
PythonOpenCVTorchVision/YOLONumPy/PandasMatplotlib

Virtual Facilitator

Jan 2023 - Jan 2024

  • Built meeting-analysis pipeline: transcription, speaker diarization, LLM-powered summarization with action-item extraction
  • Curated annotated transcript dataset; implemented precision/recall evaluation scripts
PythonWhisper/ASRpyannoteOpenAI APIFastAPIDocker

$ grep -r "skills" ./resume

// ML/AI

PythonPyTorchTorchVisionOpenCVYOLOscikit-learnNumPyPandasMatplotlibAdversarial Attacks (FGSM, PGD)Adversarial Training

// HPC/Systems

LinuxBashCUDASLURMMulti-GPU TrainingGitCI/CD

[PARSING COMPLETE] // all data extracted successfully

exit code: 0_