I'm Edgar Navarro

During my time at San Diego State University, I was not only dedicated to my academic pursuits but also actively involved in extracurricular activities that fostered my passion for technology and collaboration. I participated in the Chess Club and the Aztec Game Lab, where I had the opportunity to take on leadership roles and contribute to large-scale projects. These experiences honed my ability to work effectively in teams, communicate ideas clearly, and manage complex tasks. I gained valuable insights into the software development lifecycle, from conceptualization to deployment, and developed a strong appreciation for the importance of teamwork and effective communication in achieving shared goals.

Developed a blockchain-based couponing system for CPGs, retailers, and clearinghouses, ensuring transparent and immutable tracking of coupon creation, redemption, and reconciliation. Rigorous testing demonstrated the platform's resilience to fraudulent activities, ensuring coupon integrity. This innovation significantly reduced the risk of coupon fraud and discrepancies.

Developed and integrated a web application feature for dynamic content generation based on user-specific data. Implemented secure LinkedIn OAuth authentication and deployed RESTful APIs to fetch, analyze, and present personalized insights using OpenAI's GPT model.

Developed a blockchain-based system to enhance neural network training integrity. Conducted experiments resulting in a notable increase in model verification accuracy, exceeding 80% in a system with 50% malicious nodes, ensuring Byzantine Fault Tolerance (BFT) and resistance to Sybil attacks. The system effectively reduced the risk of model manipulation and data poisoning.

Developed a domain-specific topic modeling pipeline, enhancing model accuracy by 15% over traditional methods. Utilized tokenization, lemmatization, and stemming with Python, NLTK, and Gensim. Published findings on the comparative effectiveness of various topic modeling methods.

Re-designed a species classification system using PyTorch, achieving an 18% increase in identification accuracy. Integrated convolutional neural networks, image segmentation, and data pooling techniques. Managed and optimized real-time camera trap data, enhancing monitoring efficiency and data reliability.