Western Washington University

• Designed and built a novel adversarial framework in Python that outperformed state-of-the-art detection methods, validating results against real municipal network data with cross-dataset validation.
• Developed a novel encoding scheme, rethinking adversarial DNS Exfiltration by using combinatorial mathematics to optimize data transmission through structural patterns rather than payload manipulation.
• Implemented reinforcement learning and LSTM models to optimize system configurations and mimic realistic network behavior, maximizing throughput while evading detection.
• Engineered concurrent data pipelines for simulating, parsing, and validating large volumes of DNS records.

• Taught Computer Systems 1 lab across 4 sections (~100+ students), grading code, tests, and written assignments, and holding office hours.
• Graded assignments for CSCI 447: Operating Systems.

• Designed the frontend architecture for SPIE's Digital Library, building a React/TypeScript component library that coordinated React Islands via a Redux global store, allowing for consistent UI/UX across the platform.
• Refactored legacy model-bound views into well-structured API and Entity Controllers in C#/ASP.NET MVC.
• Redesigned the Digital Library's session system end-to-end, from frontend state to backend authentication controllers.
• Decreased page load times by 90% through Webpack bundle optimizations, lazy loading, and Redis caching.

• Taught CSCI 247: Computer Systems 1 Lab across 2 sections, grading assignments and holding office hours.

• Spearheaded backend development as the sole backend developer using Python (Django) with frontend development in TypeScript (React), integrating with Canvas LMS.
• Conducted qualitative research by interviewing faculty about grading workflows and instructor-TA communication processes.
• Designed backend architecture and RESTful API endpoints for grading, feedback, and course management operations.

• Led development of the Vulkan Synthetic Vision System for Raspberry Pi 5 using Go.
• Collaborated with a multidisciplinary research team, driving meetings and aligning technical direction with project objectives.

• Automated data processing workflows in Go and Python, with PDF processing, reducing manual review by 10 hours weekly.
• Created a duplicate-checking system that identified matching SSNs and names, saving over 200 hours of manual review over the summer and reducing human error in data verification.
• Partnered with team members to identify and automate repetitive workflows, increasing efficiency and accuracy.
• Engineered a flexible document processing framework with configuration-based rule management, enabling non-technical staff to update processing logic independently.

• Evaluated 6+ neural architectures for terrain compression, achieving 30x compression over raw elevation data.
• Built a real-time Vulkan compute shader terrain renderer in Zig with a 7-level clipmap LOD system, hitting 20,000+ FPS on desktop and 300+ FPS on a Raspberry Pi 5. Compression error is within source data noise.
• Designed a training pipeline processing Copernicus GLO-30 GeoTIFF tiles with per-channel INT8 quantization, producing models 50-75x smaller than PNG baselines for global terrain coverage.

• Designed and developed a full-stack course scheduling application serving up to 1,000 monthly users, using bitmask-based conflict detection with backtracking to generate optimized schedules based on timing, compactness, and GPA.
• Built a Go backend with SQLite persistence, a background job scheduler for automated data synchronization, and a concurrent web scraper pulling from WWU's Banner API to maintain up-to-date course catalogs.
• Developed a responsive React/TypeScript frontend with persistent client-side state, real-time course search with relevance scoring, and an interactive calendar interface for browsing and comparing generated schedules.

• Trained a DQN policy in Webots simulation for autonomous light-seeking on a 29g Crazyflie drone, using camera-based brightness detection and ToF rangefinders for obstacle avoidance.
• Deployed the trained model on-device via ONNX on the GAP8 RISC-V MCU, running real-time inference alongside a PID flight controller on the STM32 Cortex-M4 (168 MHz, 192 KB SRAM).
• Built the full sim-to-real pipeline including a Gymnasium environment with custom reward shaping, relative brightness CV, and offboard-to-onboard deployment staging.

• Built a TypeScript VSCode extension with custom lexer and abstract syntax tree parser to provide syntax highlighting and linting for KPL (Kernel Programming Language) and its compiled Blitz assembly output.
• Implemented context-aware hover documentation, 7+ diagnostic lint rules for symbol validation, and an automated Blitz assembly formatter with label-based indentation and comment alignment.
• Published to the VS Code Marketplace with automated CI/CD via GitHub Actions.

Project: Vulkan Synthetic Vision System. Faculty Spotlight