Skip to content

noaboa07/ai-projects-portfolio

Repository files navigation

Artificial Intelligence & Algorithmic Foundations

Graduate Technical Portfolio | Master of Science in AI (May 2026)

This repository serves as a comprehensive technical evidence locker for my Master’s degree research and implementation projects. It spans the full spectrum of modern AI—from the mathematical first principles of optimization to the deployment of generative diffusion models and scalable big data architectures.


📁 Repository Structure

Mathematical implementations, Optimization, and Big Data pipelines.

  • PySpark Big Data Pipeline: Distributed linear regression and clustering on large-scale Spotify networks using Spark MLlib.
  • Community Detection: Girvan-Newman edge-betweenness vs. Louvain modularity optimization on social retweet data.
  • Manual PCA & Eigendecomposition: From-scratch implementation of PCA and Minimum Distance Classifiers (NumPy only).
  • Gradient Descent Optimizer: Custom GD engine with manual partial derivative calculations for non-convex topologies.

Advanced image analysis and regularization studies.

  • Multi-Task Facial Analysis: A shared-backbone CNN for simultaneous age regression and gender classification.
  • Regularization Ablation: 10-fold cross-validation analysis of Batch Normalization and Dropout impact on generalization.
  • MNIST Heuristics: Manual feature extraction and quadrant-averaging studies for digit classification.

High-fidelity signal analysis and generative modeling for physical domains.

  • Spatio-Temporal Physical Forecasting: Multi-step forecasting using SVD for EOF dimensionality reduction and PyTorch (RNN, LSTM, Transformer).
  • Underwater Acoustic Diffusion: Using DDPM Diffusion Models and VAEs to generate synthetic sonar/ship data to train robust classifiers.
  • Acoustic Signature Classification: Comparative study of Transformers vs. VAE-based augmentation for underwater signal processing.

Autonomous reasoning, goal-oriented logic, and LLM orchestration.

  • Autonomous RAG Agent: Built a LangChain-powered intelligent agent capable of multi-tool routing, real-time math execution, and Retrieval-Augmented Generation.
  • Informed Heuristic Search: A* Pathfinding and Utility-Maximizing agents for complex grid environments.
  • Goal-Based Agents: Implementations of agents designed for environment-specific utility maximization.

Sequence modeling, LLM orchestration, and graph-based NLP.

  • Biomedical Abstract Classification: Bidirectional LSTM vs. SimpleRNN comparative study on technical, domain-specific text.
  • RAG & Text Summarization: Implementation of Retrieval-Augmented Generation systems for automated document synthesis.
  • Social Link Prediction: Topological feature engineering (Adamic-Adar, Jaccard) for predicting edges in social graphs.

Reward-based optimization, continuous control, and Markov Decision Processes.

  • Continuous Control via DDPG: Deep RL Actor-Critic optimization for the OpenAI Gym Pendulum-v1 environment.
  • Q-Learning & Multi-Armed Bandits: Implementation of tabular Q-learning and Epsilon-Greedy reward strategies.
  • Policy Optimization: Benchmarking learning agents against stochastic environmental constraints.

Foundational search logic and pathfinding.

  • Uninformed Search: BFS and DFS implementations for automated maze solving.
  • Stochastic Search: Simulated Annealing for constraint satisfaction problems (N-Queens).

💼 Professional Experience

  • Software Engineer | Sea Turtle Oversight Protection (Jan 2025 – Dec 2025)
    • Engineered and maintained client-facing software solutions, ensuring high reliability and performance for environmental oversight operations.

🛠️ Technical Stack

  • Languages: Python (Advanced), TypeScript, C++, SQL
  • AI/ML Frameworks: PyTorch, TensorFlow/Keras, PySpark, Scikit-Learn, LangChain
  • Web & Cloud: Next.js 14, React, Firebase, REST APIs
  • Specialized Libraries: NumPy, Pandas, Matplotlib, NetworkX, SciPy, torchaudio, diffusers
  • Core Competencies: Deep Reinforcement Learning (DDPG), Signal Processing (SVD/EOF), Computer Vision (CNN/VAE), Generative AI (Diffusion), NLP (RAG/Transformers), Big Data (Spark), Full-Stack Development.

🎓 Education & Career Objectives

  • M.S. in Artificial Intelligence | Florida Atlantic University (Expected May 2026)
    • GPA: 4.0 | President’s Honor List (7 Semesters)
  • B.S. in Computer Science (Minor: Cybersecurity) | Florida Atlantic University
  • Location: Colorado Springs, CO
  • Objective: Seeking Graduate/Junior roles in AI Engineering, Software Engineering, or Signal Processing.
  • Eligibility: U.S. Citizen | Clearance Eligible.

Noah Russell LinkedIn | Email

About

Master’s Level AI & Machine Learning Portfolio. Featuring custom Transformers, PyTorch Diffusion models, Spatio-Temporal physical forecasting, Deep Reinforcement Learning (DDPG), and Big Data pipelines (PySpark). Specialized in Signal Processing, Computer Vision, and Algorithmic Foundations for Aerospace and Defense applications.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors