- Player Name : Melvyn Avoa
- Team : University of Ottawa
- Position (Program): 3rd Year, Bachelor of Computer Science
- Strengths (Specializations): 🛡️ Cybersecurity, ☁️ Cloud Computing, and 🧠 AI/Machine Learning.
- Draft Status (Objective): Actively seeking SUMMER 2026 internship opportunities (May - August).
- Game Plan (Motivation): Eager to leverage strong theoretical knowledge into real-world projects and secure a high-value co-op experience.
##1 : 🛡️ GabrielAI: Machine Learning Phishing Detection Engine
Live Demo : https://gabrielaiv1.streamlit.app/
Goal: Develop a real-time detection tool capable of identifying "zero-day" phishing attempts by analyzing URL structure and lexical features, rather than relying on traditional static blacklists. Status: v1.0 Completed & Deployed (November 2025). Optimization and API integration planned for Q1 2026. Targeted Technologies: Python 3.10, Scikit-Learn (Random Forest), Pandas/NumPy, Streamlit Cloud, Git/GitHub. Demonstrates: Applied Machine Learning: Implementation of Feature Engineering to translate raw URLs into numerical vectors for threat analysis. Threat Intelligence: Understanding of common obfuscation techniques (IP usage, URL depth, character masking) used in social engineering attacks. Full-Stack Development: Built an end-to-end pipeline from data extraction to a functional web interface. Research & Methodology: Ability to evaluate model precision and minimize false positives in a security context.
Repository Link: https://github.com/MelvynAv/GabrielAi
2. ☁️ Network - Repository Link: https://github.com/MelvynAv/signal-stream
Demo Link: https://signal-stream-demo.onrender.com Real-Time Network Telemetry Dashboard
- Tech: Python (FastAPI), React, WebSockets, AsyncIO, AWS EC2.
- About: A full-stack monitoring tool that streams network latency metrics in real-time (500ms intervals). Features an automated Anomaly Detection Engine to flag network degradation instantly.
- Status: 🟢 Live / Open Source
- Goal: To bridge the gap between raw computer vision and high-level semantic reasoning by enabling autonomous systems to "describe" and "understand" the physical spaces they occupy.
- Status: Completed and deployed live (Q1 2026)
- Targeted Technologies: Python, FastAPI, YOLOv8 (Computer Vision), Groq LPU Inference, Llama 3.3 (LLM), and Render (Cloud Infrastructure).
- Demonstrates: Real-time object detection, spatial coordinate extraction, 3D context-aware reasoning, and secure cloud deployment (DevSecOps).
- Repository Link: https://github.com/MelvynAv/physicast
- Live demo : https://physicast.onrender.com
- 📄Resume/CV: View My Resume
- LinkedIn: My LinkedIn profile
- Email: [email protected]
