Building machine learning products that feel thoughtfully engineered.

DA-IICT

• Engineered a reproducible RL research pipeline using Docker, reducing model evaluation time by 40% and boosting accuracy by 20%.
• Developed robust experiment harnesses for hyperparameter optimization and ablation studies, integrating Weights & Biases for metric visualization.
• Produced high-quality, reusable code modules to support ongoing and future studies.

Independent Research

• Architected a novel reasoning framework augmenting LLMs with Proximal Policy Optimization (PPO) and RAG-based grounding.
• Demonstrated a 60% relative improvement on the GSM8K benchmark over baseline models.
• Developed a complete research codebase, including an evaluation harness and interpretability tools.

Gannon University

• Constructed a hybrid model achieving 97% accuracy on a dataset of 100,000+ APKs.
• Validated model robustness against adversarial attacks and utilized SHAP for post-hoc interpretability.
• Accepted for presentation at the Microsoft Future Tech Conference (Nov 2025).

Grew community to 50+ members; organized AI/ML workshops, hackathons, and mentored peers on research methods.

Led a team of four to win a university-level cybersecurity competition focused on reverse engineering, cryptography, and exploitation.

Maintained 95% lab uptime and reduced average IT ticket resolution time by 30%.