
Khondoker Sazzad Sunfi
Data Science | Machine Learning | AI
"I build intelligent systems and scalable data architectures that tackle complex challenges. With deep expertise in machine learning and AI, I turn raw data into powerful actionable insights and production-ready AI products that deliver real value."
Contact
Education
United International University
Bachelor of Science
2023 — Present
St. Gregory's High School and College
Higher Secondary Certificate (H.S.C)
2019 — 2021
Languages
Khondoker Sazzad Sunfi
Data Science | Machine Learning | AI
Professional Summary
I’m a passionate builder of intelligent systems who loves turning raw, chaotic data into clear, actionable insights and powerful AI products. With a deep focus on scalable data architectures and robust machine learning solutions, I thrive on solving complex problems that actually matter. Whether it’s designing efficient data pipelines, training models that perform in the real world, or creating AI systems that scale gracefully — I’m driven by the excitement of watching data come alive and deliver meaningful impact. Constantly curious and hands-on, I enjoy crafting end-to-end AI solutions that are not just technically strong, but genuinely valuable.
Technical Skills
| Category | Technologies |
|---|---|
| AI & Machine Learning | Machine Learning,Time Series Analysis,LLM,NLP |
| Data Science & Analytics | Python,Web Scraping,MySQL,PostgreSQL,Pandas,NumPy,Seaborn,Plotly,Streamlit,Data Visualization |
| Full-Stack Development | Next.js,JavaScript,TypeScript,React,Tailwind CSS,Framer Motion,Vite |
| Backend & Engineering | FastAPI,Fiber,Pydantic,SQLAlchemy,REST API,Uvicorn,Automation (n8n) |
| Cloud & MLOps | Git,GitHub,Docker,Sanity.io,Supabase,Vercel |
Experience
Student Developer
Nov 2024 —Jan 2025Key Responsibilities:
- •Developed an award-winning full-stack Streamlit web application for campus lost and found item management
- •Designed and implemented scalable data architecture using Object-Oriented Programming (OOP) principles
- •Built interactive dashboards with real-time data visualization for students and administrators
- •Implemented core features including class-based authentication, item reporting, tracking, and image upload system
- •Transformed raw CSV data into actionable insights using Pandas, Matplotlib, Plotly, and Seaborn
Key Achievements:
- ★🏆Secured 1st Runner Up position in the OOP Course Project Competition at United International University
- ★Successfully built and delivered a production-ready Lost and Found Management System serving the university campus community
- ★Created end-to-end intelligent data application that transformed raw CSV data into actionable insights through interactive visualizations
- ★Demonstrated strong problem-solving and software development skills by completing a robust, user-friendly system in a team environment
Projects
UIU Lost and Found Management System
- •Award-winning web application built to efficiently manage lost and found items on the United International University campus. The system allows students and administrators to report, track, and manage items with interactive dashboards and real-time data visualization.
Bangladesh Data Science Job Market Intelligence Pipeline
- •Automated Scraping: Developed a robust Python scraper with BeautifulSoup and Requests to ethically harvest tech job listings from LinkedIn, incorporating random delays and user-agent rotation to avoid rate limits.
- •ETL & Data Wrangling: Engineered pipelines using Pandas and Regex to filter data-related posts, drop duplicates, clean geographical/temporal fields, and extract structured tech-skills from unstructured descriptions.
- •Batch Integration: Built an automated batch utility to merge multiple fragmented raw CSV files (batches A through P) into a single unified, query-ready 6MB+ master repository.
- •Interactive BI Dashboard: Designed a real-time analytics web dashboard using Streamlit to visualize job market velocity, top hiring companies, city-wise distributions, and top skill trends via Seaborn and WordCloud
UIU Shuttleflow: Automated Campus Transit & Fleet Management Platform
- •Decoupled Architecture: Engineered a full-stack campus transit ecosystem for UIU using a customized Streamlit frontend and a high-performance MySQL relational database backend.
- •Relational Database Design: Built a robust MySQL schema enforcing strict foreign key constraints and optimized SQL JOIN views (schedule_overview), guaranteeing 100% referential integrity.
- •Commuter Portal & Feedback Loop: Created an interactive student panel to track departure schedules and lodge transport complaints, reducing administrative resolution time by 65%.
- •Fleet Analytics & CRUD APIs: Developed an admin command center with dynamic Plotly charts for real-time fleet tracking, secure transactional CRUD APIs, and one-click enterprise CSV data exports.
