Skip to content
View florykhan's full-sized avatar

Block or report florykhan

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
florykhan/README.md
Banner

Hello, I'm Ilian πŸ‘‹

CS & Math @ SFU Β· Data, ML, Software


Portfolio LinkedIn Email Resume


πŸ”­ Current focus

Building software, machine learning, and data analytics projects. Interested in applied ML, data engineering, and clean systems.

  • Data β€” Cleaning, analysis, and visualization (SQL, Python, Tableau); dashboards and BI for exploratory and market-style work.
  • ML β€” End-to-end pipelines, feature engineering, interpretable models (e.g. CatBoost), and evaluation (ROC AUC, calibration, cross-validation); regression and classification.
  • Software β€” Full-stack web apps (React, Flask), APIs, geospatial/maps (e.g. Leaflet), and LLM or voice-driven UIs.

πŸ“ Featured projects

Data Analytics & BI

Project Description Tech
customer-churn-dashboard End-to-end churn analytics: SQL β†’ R β†’ Tableau pipeline (Telco dataset). Executive dashboard for churn patterns, high-risk segments, revenue at risk. SQL SQLite R dplyr tidyr Tableau
airbnb-market-analysis Vancouver Airbnb: raw CSV β†’ Python ETL β†’ SQLite, SQL aggregation, interactive Tableau dashboard for pricing, neighbourhood, and room-type insights. Python pandas SQLite SQL Tableau Jupyter

Machine Learning Projects

Project Description Tech
graduate-underemployment-prediction Full ML pipeline for predicting overqualification (underemployment) from NGS hiring data. CatBoost, stratified K-fold CV, feature importance, optional SHAP. SFU ML Hackathon. Python CatBoost scikit-learn pandas SHAP Jupyter
house-price-ml-v2 Modular ML pipeline for California median house price prediction: feature engineering, OLS/Ridge/Lasso, custom Gradient Descent, 5-fold CV, hyperparameter tuning. Python scikit-learn pandas NumPy Matplotlib Jupyter

Full-Stack Applications

Project Description Tech
FutureSpend Calendar-driven personal finance: predicts upcoming spend from events, category-based pipeline, dashboard with 7-day forecast, challenges, AI coach (Gemini). Next.js + FastAPI. SFU/RBC hackathon. Next.js TypeScript Tailwind FastAPI Recharts Gemini
TelusGuardAI AI-powered network impact analyzer: multi-agent orchestration (Gemma, GPT), NL queries, web/weather intel, geospatial impact. Interactive Leaflet map, severity and confidence scoring. React Vite Leaflet Flask LLM
GymWhisper Voice-powered workout tracking: hands-free logging via speech β†’ Gemini API parses to structured exercises; client-side, bilingual (EN/FR). React Gemini speech-recognition FramerMotion

Independent Projects

Project Description Tech
florykhan.github.io Static portfolio: projects, experience, education, contact. Vanilla HTML, CSS, JavaScript. GitHub Pages + CI/CD. HTML CSS JavaScript

🎯 Tech stack

From featured projects above.

Languages
Python R SQL JavaScript TypeScript HTML CSS

ML & data
pandas NumPy scikit-learn CatBoost Matplotlib Seaborn Jupyter Tableau SQLite SHAP

Web & tools
React Next.js Vite Flask FastAPI Leaflet Tailwind CSS LLM


πŸ“Š Stats

GitHub streak

πŸ“… Activity

GitHub contribution graph


🀝 Let's connect

Reach out for collaboration, internships, or just to say hi.

Portfolio Β· LinkedIn Β· Email Β· Resume (PDF)

Pinned Loading

  1. FutureSpend FutureSpend Public

    AI-powered personal finance dashboard that predicts upcoming spending from calendar events and transaction data. Built with Next.js, TypeScript, Tailwind, and a Python API deployed on Render.

    TypeScript 1 1

  2. customer-churn-dashboard customer-churn-dashboard Public

    End-to-end analytics pipeline (SQL β†’ R β†’ Tableau) for telco customer churn: KPIs, segment breakdowns, and revenue at risk.

    R

  3. graduate-underemployment-prediction graduate-underemployment-prediction Public

    End-to-end ML pipeline for predicting graduate underemployment from NGS 2020 survey data, including EDA, feature engineering, categorical-aware modeling with CatBoost, and ROC AUC validation.

    Jupyter Notebook

  4. airbnb-market-analysis airbnb-market-analysis Public

    End-to-end data analysis of Vancouver Airbnb listings using SQL, Python, and Tableau dashboards.

    Jupyter Notebook

  5. TelusGuardAI TelusGuardAI Public

    AI-powered network impact analyzer. Natural-language queries β†’ multi-agent pipeline β†’ geospatial impact + interactive Leaflet map. Flask + React.

    JavaScript

  6. house-price-ml-v2 house-price-ml-v2 Public

    End-to-end ML pipeline for California house price prediction. Features engineered data, OLS/Ridge/Lasso models, custom Gradient Descent, cross-validation, hyperparameter tuning, and a full training…

    Jupyter Notebook