Portfolio · LinkedIn · Email

• End-to-end security tools – supply chain attack detection, encrypted traffic monitoring, malware scanning.
• ML-powered developer tooling – graph neural networks over package ecosystems, intelligent document parsing.
• Full-stack products – TypeScript/React/Next.js frontends backed by robust APIs in Node.js or C#/.NET.
• Systems that turn manual, error-prone workflows into repeatable pipelines with tests, metrics, and docs.

I’m currently finishing my BSc in Computer Science at the British Columbia Institute of Technology (BCIT) and actively working on security + ML heavy projects, alongside real-world web apps.

Repo: Sentinel

Graph-neural-network-based tooling to detect suspicious packages and dependencies in the npm ecosystem.

• Builds a dependency graph over npm packages and release metadata, then uses a GNN to flag anomalous nodes.
• Pipeline in Python with Neo4j (graph store), PyTorch Geometric (GNN), and ONNX (for portable inference).
• Targets patterns like supply-chain attacks: sudden maintainer changes, typosquats, unusual fan-out/fan-in, and odd versioning behaviour.

Repo: CoDefend

A security-first Chrome extension that scans downloads before they hit disk.

• Intercepts browser downloads and forwards files/URLs to a Node.js backend for scanning with multiple AV and reputation engines.
• Designed to decentralize malware analysis: the extension stays lightweight while the backend fans out to external scanners.
• Built with JavaScript/TypeScript, Node, Express, and REST APIs, with a strong focus on user safety and clear risk reporting.

Early-stage B2B SaaS aimed at property managers, landlords, and regulators.

• Extracts structured data from leases, invoices, compliance forms, and scanned PDFs using OCR + LLMs.
• Uses tools like Surya-OCR, Azure Document Intelligence, and a vector database (e.g., Qdrant) for semantic search over documents.
• Focused on real-world pain points: compliance checks, expiry tracking, and reducing the time required to find critical clauses.

Repo: Melloww

A mobile-first mental health journaling + social platform built with a small team.

• Led both design and development in a semi-agile process; guided junior teammates through bugs, PRs, and reviews.
• Features journaling, mood tracking, and community-style interactions.
• Stack: TypeScript, Next.js, Tailwind CSS, shadcn/ui, Node.js, Express, MongoDB.

Repo: EngineerInTech

A platform for engineers to meet by teaching each other skills instead of swiping on job posts.

• Full-stack TypeScript project with a separate client and server codebase.
• Matching flow focuses on skills you’d like to teach and skills you’d like to learn.
• Stack: TypeScript SPA client + Node backend; deployed at engineerin.tech.

Repo: NoteTed

Fast note-taking app with Next.js SSR and Firebase.

• Users can register/login and create notes that sync across devices.
• Achieves instant load times via server-side rendering and efficient data fetching.
• Stack: Next.js, React, MUI, Firebase Auth, Firebase Firestore.

Languages
TypeScript · C# · Python · JavaScript · C/C++ · Java · SQL

Backend
Node.js · Express · FastAPI · .NET · REST APIs · WebSockets

Frontend & Mobile
React · Next.js (App Router) · React Native · Tailwind CSS · shadcn/ui · MUI

Data, ML & Storage
PyTorch Geometric · XGBoost · Neo4j · SQL Server · MongoDB · Firebase (Auth/Firestore) · Vector DBs (Qdrant)

DevOps & Tooling
Docker · GitHub Actions · Azure DevOps · Vercel · Cloudflare · npm/yarn/pnpm · Jest · React Testing Library · pytest

• EventPrompt – Discord bot for timed reminders using Discord.js, Express, Firebase Firestore.
• Anonymous Confession App – Android app with Firebase Auth + Cloud Firestore and privacy-preserving login.
• UBC Solar Website – Mobile-first React site for the UBC Solar team, built with HTML, CSS, JavaScript, React.

I’m excited about roles where I can:

• Work close to security, ML, or developer tooling.
• Own features end-to-end – from design docs and data pipelines to tests, dashboards, and production readiness.
• Collaborate with strong engineers who care about code quality, observability, and long-term maintainability.

If you’re building something in this space and need someone who loves TypeScript, C#/.NET, Python, ML, and security, I’d love to chat.