I'm a software engineer with a strong foundation in computer science, statistics, and mathematics, focused on building practical systems that work end-to-end — from data and models to clean, usable products.

🎓 B.Sc. Computer Science & Statistics, University of Toronto (Math minor)
🧠 Background in ML, statistical modeling, and systems design
⚙️ Comfortable owning projects from architecture → implementation → deployment

I care about clarity, focus, and execution — shipping useful software, not hype.

📌 LinkedIn · 🌐 Portfolio · 📚 Goodreads

• Applied AI systems (LLMs, RAG pipelines, anomaly detection, inference workflows)
• Backend-leaning full-stack engineering
• Automation & developer tooling that saves real time
• Writing maintainable systems that scale beyond demos

I prefer deep work on fewer things over juggling many shallow projects.

Security Anomaly Detection Pipeline (Maritime Financial Group)

• End-to-end firewall and VPN anomaly detection processing 100k+ log events weekly
• Hybrid ML pipeline: rules engine + Isolation Forest + autoencoder over 5-minute windows
• Real-time Next.js dashboard with grouped alerts, drill-down views, and trend analytics
• Reduced manual analysis time by 70%+ through automated risk scoring

AI Mock Interview Platform

• On-demand AI voice interviews with structured written feedback (Next.js · Vapi AI · Google Gemini · Firebase)
• Designed objective evaluation schemas validated across 100+ sessions
• Built and deployed end-to-end as a solo project

Inbox Copilot

• Keyboard-first email client with real-time sync and multi-account support
• AI copilot drafts replies and answers inbox questions using RAG over prior threads (LangChain · Pinecone)
• Subscription SaaS with Stripe payments, feature gating, and Clerk auth

AI Developer Collaboration Platform

• Tool for querying and understanding large GitHub repositories
• Production-grade RAG pipeline over real codebases with source-grounded, verifiable answers
• Multi-user workflows with access control, usage limits, and credit-based billing

Hot Hand Insight — Bayesian NBA Analysis

• Analyzed the "Hot Hand Fallacy" across 15 NBA players using Bayesian regression in PyMC
• Compared Multivariate Normal, Horseshoe, and Spike-and-Slab priors; reduced model divergences by 20%
• Presented to class and the UofT Statistics department head

Languages
Python · TypeScript · JavaScript · Java · R · C · SQL

AI / ML
LLM APIs (OpenAI / Gemini) · LangChain · Hugging Face · PyTorch · scikit-learn · PyMC

Frameworks
React · Next.js · Node.js · Express · FastAPI · Spring Boot

Data & Infrastructure
PostgreSQL · MongoDB · Prisma · Docker · AWS · Firebase · Pinecone · Vercel

• Software Engineering (backend-leaning, full-stack)
• ML / AI Engineering (applied LLM systems, RAG, anomaly detection)
• Developer tools & internal platforms

I'm especially interested in roles where I can own systems end-to-end and grow fast.