|

What I've Worked On

Kairo — AI University Assistant
01

Kairo — AI University Assistant

An AI-powered platform for uOttawa students to automate schedule planning, get real-time course and professor insights, and ask natural-language questions about programs and prerequisites.

  • Real-time university data at scale with resilient web scraping (dynamic selectors, iframe handling, rate limiting, deduplication)
  • Production-grade system: multi-service architecture, automated pipelines, type-safe full-stack, responsive animated UI
live courses indexed
10,000+
subjects
160+
professors
1200+
PythonTypeScriptDjangoReactNode.jsPostgreSQLOpenAI
Data Collection PipelineScheduleCourse OfferingsPuppeteer • PlaywrightCatalogueDescriptionsProf DirectoryuOttawa OfficialPipelineParse • CleanJSON Filescourses.json15k Courses160 Subjects1,200+ Profs200 Programs3 Terms6 Active
02

Your Scrapers Overview

Multiple Headless scrapers that collect uOttawa course data across terms and expose structured, real-time availability.

total courses
15k+
subjects
160+
programs
200+
terms
F25, W26, S/S25
TypeScriptPuppeteerPlaywrightExpress.js
LeetHubAuto-sync your LeetCode solutions to GitHubStatus:ConnectedRepository:oumizumi/leetcode-solutionsBranch:mainPush Current PageSTATISTICS24TOTAL SOLVED24EASY0MEDIUM0HARDRECENT ACTIVITYRefreshPushed: Two SumEasy • Array • Hash Table2d ago
03

LeetHub

A Chrome extension that automatically synchronizes accepted LeetCode solutions to GitHub, enabling me to maintain a well-organized portfolio of their coding practice with real-time detection, smart organization, and zero-friction backup.

  • Real-time detection of accepted solutions with automatic GitHub push within 2 seconds
  • Modular architecture: service workers, content scripts, GitHub API integration with retry logic and error handling
Auto-sync time
<2s
Core Modules
6
Success Rate
99%+
JavaScriptChrome APIGitHub APIManifest V3REST API
Premier League Score Predictor
04

Premier League Score Predictor

Machine learning model that predicts football match scores using Random Forest regression; trained on 137+ real matches with rolling weekly ingestion and automated retraining, achieving ~0.5 goals MAE accuracy.

  • I love football - Eric Cantona
Chelsea winning the league
<1%
Chelsea winning the UCL
<1%
Charity FC - making mid-table teams look elite
∞%
PythonNumpyScikit-learnPickle

A Bit About Me

i enjoy creating software thats clean, reliable, and easy to use. i usually spend my free time watching football/soccer, spending with with my family and friends, or out in nature.

Get in touch

Tech Stack

  • JavaScript
  • TypeScript
  • Python
  • Java
  • C++
  • NumPy
  • Tensorflow
  • Scikit-learn
  • React
  • Django
  • Node.js
  • Tailwind CSS
  • PostgreSQL
  • OpenAI
  • Puppeteer
  • Playwright
  • REST API
  • GraphQL
  • Git
  • Docker
  • AWS
  • MongoDB
  • Vercel
  • Supabase

Get in touch

I'm always open to hearing about new projects and opportunities. Whether you have a question or just want to say hello, I'll get back to you as soon as possible.

oghar074@uottawa.ca
LocationOttawa, ON