I'm a Computer Science student at Western University, based in Toronto, Ontario, with hands-on experience building full-stack, cloud, hardware, and AI-enabled software systems.
- 🔭 I’m currently working as a Software Developer at EMRC Lab, where I build backend APIs, improve device-status reliability, and support field-device research workflows.
- 🚀 I’m also the founder of SigFig AI Security Services, an AI-enabled gym access security startup focused on tailgating detection, fraud visibility, and operational analytics.
- 🧠 I’ve built production systems including a QR tracking SaaS used by 10,000+ users, geoscience research platforms, automation tools, and open-source machine learning runtime software.
- 📚 I’m currently learning more about AI systems, product management, embedded hardware, cloud infrastructure, and scalable backend architecture.
- ⚡ In my free time, I like building startup ideas, experimenting with hardware, improving my portfolio, and creating tools that solve real-world problems.
AI-enabled gym access security startup focused on helping gyms detect tailgating, reduce unpaid entry, and understand access-risk patterns through operational analytics. I founded the startup and have been leading product strategy, market validation, customer discovery, and technical prototyping.
Key work:
- Built a Wi-Fi-connected hardware prototype using C++, ESP32, and PN532 NFC for real-world access verification workflows.
- Designed the core tailgating detection concept around suspicious multi-person entry events at gym access points.
- Planned dashboard analytics for peak-risk hours, repeat access patterns, incident frequency, and manager review workflows.
- Developed the product vision around timestamped evidence clips, access-card verification, and operational fraud visibility.
Open-source machine learning runtime work focused on running language-detection models efficiently in JVM environments. My contribution focused on the Java application layer, model execution pipeline, preprocessing, packaging, and validation.
Key work:
- Built the Java 17 application layer for a language-detection CLI.
- Integrated DJL + ONNX Runtime to load ONNX artifacts and run inference.
- Implemented dual preprocessing inputs using TF-IDF and token-sequence pipelines.
- Returned structured JSON prediction outputs with language-label probabilities.
- Added self-test validation to confirm model loading, preprocessing, and inference behavior.
Our Changes »
Upstream Repository »
Production full-stack QR tracking SaaS used by 10,000+ users, built with React, Node.js, Express.js, Vercel, AWS DynamoDB, Stripe, and authenticated analytics. The platform supports payment-gated access, multi-tenant data isolation, redirect analytics, dashboards, heatmaps, and exports.
Key work:
- Built scan analytics behind QR redirect flows, capturing geo, device, OS, and browser data.
- Delivered time-series dashboards, heatmaps, and exportable analytics for business users.
- Integrated Stripe Checkout/Webhooks for subscription-gated access.
- Added email verification, password reset flows, rate limiting, CSP, HSTS, and other security headers.
Research platform modernization project for geoscience and field-data workflows at Western University. I worked as Product Owner and Software Developer, supporting a platform used to record, search, and manage geospatial research data.
Key work:
- Owned phased delivery across 10 releases, coordinating priorities, stakeholder feedback, and release documentation.
- Built an Express.js REST API backed by PostgreSQL for geospatial research records.
- Added JWT authentication, bcrypt password security, CORS, and secure file-upload support.
- Supported Google Maps-based visualization and search workflows for expedition and crater-impact research data.
At HackWestern X, I led a team to build Attentify, a productivity web app and Chrome extension tailored for the ADHD community. We collected feedback from over 150 people through a Reddit poll to shape the product around real user needs.
Key features:
- Pomodoro timer for structured focus sessions.
- Google Calendar integration for planning and scheduling.
- Custom chatbot powered by the OpenAI API.
- Floating tab-specific notepad for lightweight note-taking while browsing.
- Chrome extension experience designed around distraction reduction and accessibility.
Minimalistic web clock built with HTML, CSS, and JavaScript. The app focuses on readability, clean design, and real-time visual changes based on the current time.
Key features:
- Real-time digital clock display.
- Smooth background color transitions based on time of day.
- Optional background image changes for different times.
- Simple, readable, responsive interface.
Python command-line automation tool developed at Carolinian Canada Coalition to extract, clean, and analyze large-scale Facebook post data for research workflows.
Impact:
- Automated analysis across 23,000+ Facebook posts.
- Added hash-based duplication protection to reduce repeated data issues.
- Saved 385+ hours of manual work and improved research efficiency.
- Supported research work connected to a scientific journal.
ARM7 Assembly project that validates Universal Product Codes using low-level arithmetic and control-flow logic. This project helped strengthen my understanding of computer architecture, assembly programming, registers, memory, and branch-based validation.
Key work:
- Implemented UPC validation logic in ARM7 Assembly.
- Practiced register-level operations, branching, and arithmetic checks.
- Built a clearer understanding of low-level computation and hardware-near programming.
Second-place winning IveyHacks 2023 project built for sponsor Cyder. Our team developed and pitched a mobile app prototype focused on data privacy, business value, and user experience.
Key work:
- Led scoping, backlog prioritization, and early product planning.
- Created Figma wireframes for the mobile MVP.
- Helped pitch the product to CEOs, technical leads, marketing leads, and hackathon participants.
- Combined product thinking, business strategy, and technical feasibility into a competition-ready MVP.
Small text-based Python adventure game with multiple endings, branching choices, and a hidden secret ending that rewards attention to detail.
Key features:
- Written in Python.
- Includes 2 main endings and 1 secret ending.
- Uses conditional logic, branching paths, and player-choice flow.
Java program that creates a 5x5 grid, places letters dynamically, and uses randomized direction logic to fill the board. This project demonstrates grid-based programming, loops, conditionals, and randomized placement logic.
Key features:
- Built in Java.
- Uses a 5x5 two-dimensional grid.
- Places letters using random starting positions and directional movement.
- Strengthened fundamentals in arrays, control flow, and algorithmic thinking.
- Redesigned a core device-status system at EMRC Lab, improving reporting accuracy and reducing false offline/counting states.
- Built and deployed a production QR tracking SaaS with Stripe, authenticated analytics, dashboards, heatmaps, and exports.
- Owned phased delivery across 10 releases for GeoNotebook, coordinating priorities, release documentation, and stakeholder alignment.
- Developed 120+ WCAG-compliant Drupal 10 pages and automation tools at Carolinian Canada Coalition.
- Built hardware and software prototypes using ESP32, PN532 NFC, C++, Java, Python, React, Angular, Node.js, Express.js, PostgreSQL, DynamoDB, and AWS.
- 🥈 2nd Place - IveyHacks Case Competition
- 🎓 Dean’s Honor List - Western University
- 💰 $3,500 Scholarship of Distinction
- 🧠 Founder - SigFig AI Security Services
- 🗣️ DECA Regional Finalist
- 🤝 Vice President of Events - TMA Western
