~/nathan

building practical AI systems

session://projects

$ cat projects.txt

Selected technical projects I want to feature right now.

$ ls projects/

  • Fraud Ring Investigator Arena

    Built a compact OpenEnv benchmark for sequential fraud investigation, where agents inspect hidden financial networks, manage limited budget, place interventions, and reason about delayed outcomes across procedural cases.

    PythonOpenEnvRL Benchmark
    Open project
  • Implementation of Custom Deep Learning Architectures for Computer Vision

    Built a broad computer vision portfolio spanning neural style transfer, YOLOv2 vehicle detection, FaceNet-style recognition, U-Net segmentation, ResNet50 classification, and CNN-based visual classifiers.

    PythonDeep LearningComputer Vision
    Open project
  • Foundational Machine Learning and Statistical Modeling

    Implemented a wide range of statistical and machine learning methods from scratch, including Bayes classifiers, Bayesian regression, clustering, ensemble methods, EM for mixture models, PCA plus k-NN, and regularized regression.

    PythonStatistical MLPattern Recognition
    Open project