NET01
DefenceNet
Problem. Phishing links are hard to spot, so users need protection before they click.
Contribution. Built and improved ML workflows for URL classification and real-time threat scoring, exposing inference through FastAPI services.
Software Engineer by craft.
Building intelligent AI systems.
Obsessed with how systems work under the hood, and how AI makes them smarter. I design, wire, and ship intelligent applications built for real-world traffic.
Software Engineering Technology at Centennial College, now in my third year.
Phishing detection at Datacove.ai, NLP skill matching, and REST APIs in Python.
Experiment, break it, debug it, improve it. I learn by shipping things that work.
Problem. Phishing links are hard to spot, so users need protection before they click.
Contribution. Built and improved ML workflows for URL classification and real-time threat scoring, exposing inference through FastAPI services.
Problem. Resume screening is slow and keyword-based; matching should understand skills.
Contribution. NLP skill-matching system comparing resumes with job descriptions using TF-IDF and cosine similarity, with FastAPI APIs behind React.
Problem. Clinics juggle patients, appointments and billing across disconnected tools.
Contribution. Designed the REST API architecture for appointments, billing, patient and review management, with a MongoDB data model.
scikit-learn, PyTorch, NLP, spaCy
TensorFlow, TF-IDF, Pandas, NumPy, feature engineering, model evaluation
Python / Java, JavaScript, C#, Kotlin, HTML, CSS
FastAPI, REST APIs / Spring Boot, Node.js, Express, MVC
AWS (EC2, S3, CLI) / Linux, Ansible, Render, Git, GitHub
PostgreSQL / MongoDB, MySQL, DynamoDB
› inference.run()
✓ FastAPI: 200 OK
↳ latency: 38ms
↳ model_eval: PASS
Data structures, algorithms / OOP, system design, cybersecurity
Datacove.ai · Toronto
DefenceNet — AI-powered anti-phishing platform
Centennial College — Advanced Diploma