About Me
Data Science student at the University of Waterloo with an interest for machine learning, web development, and solving complex problems.
Honours Data Science, BMath
University of Waterloo
Sep. 2024 - May 2029 (Expected)
Relevant Coursework
Activities
- Data Science Club
- Applied Math, C&O Club
- Various Hackathons + Datathons
Experience
Machine Learning Engineer (Incoming)
IncomingShopify
- >Incoming Machine Learning Engineer joining Shopify for Fall 2026
Founding Engineer
Magnolia Education
- >Built a real-time collaborative math whiteboard on Next.js/Vercel using Konva.js and Yjs CRDT websocket sync for live tutoring, replacing a $6K/year commercial vendor with $600/year owned infrastructure
- >Building a Socratic Claude API tutor with vector retrieval over a session corpus, plus an OCR-driven worksheet-to-LaTeX pipeline producing aligned-environment math blocks and practice quizzes for tutors
- >Building an end-to-end student onboarding pipeline (TutorBird API + Kit email + Twilio SMS + Next.js portal) to automate new-student intake for the company's 80+ active students
AI / Software Engineer
Practice Better
- >Designed a reusable semantic search pipeline (MongoDB) shipped across three production surfaces: a 3,000+ ticket support knowledge base, the practitioner help centre, and a context-retrieval tool
- >Built an LLM-powered auto-charting pipeline that generates patient charts from session notes via a chain handler pattern, cutting charting time from 12 minutes to 2 minutes per chart
- >Developed an LLM-powered ICD-10 diagnosis code classifier across 70,000+ diagnostic codes, suggesting codes from session note summaries at an 83% F1 score
Data Analyst, Supply Chain
Celestica
- >Trained a generative AI/OCR model to extract product specifications — case size, resistance, and maximum voltage — from technical drawings at a 97% F1 score, eliminating manual data entry
- >Built an APIFY-driven scrape + Python/Pandas ETL pipeline; raised benchmarked-part coverage by 18% and total dollar value by 22%
Machine Learning Research Assistant
Fluid Flow Physics Group
- >Built a machine-learning regression pipeline mapping laryngeal muscle activations to fundamental frequency and sound pressure level in vocal fold biomechanics, achieving R² = 0.91 on simulation-based acoustic predictions
- >Authored Python scripts to transform vector data from VTU-formatted files into Pandas data frames then export as CSV for further analysis
Memory Archive
Recovered data fragments from previous constructs. Each memory represents a unique solution to complex problems.
Argus — Arctic Defence Platform
Full-stack geospatial dark-vessel detection platform that correlates SAR satellite imagery with live AIS tracking to surface suspicious ships in Arctic waters. Co-Founder.
posture-probe
Playwright + Claude agent that drives the GitHub admin console end-to-end and audits live configuration against a CIS-style security benchmark, emitting structured PASS / FAIL / UNKNOWN findings with verbatim evidence quotes
Red Light Detection — Raspberry Pi CV
On-device YOLOv5-nano traffic-light classifier running at 12+ FPS on a Raspberry Pi 4 with Pi Camera, hitting 94% mAP on a custom 3,000-image dataset
SleepyPlays
Automated YouTube Shorts content creation system using AI for story generation, voice synthesis, and scheduled video publishing